No, seriously. It really is. Those of you who are too young to remember the heyday of Computer Shopper won't get it (probably) but that magazine was ... well, as the article says, practically a "bible" to some of us. There was NOTHING like finding the new issue on the news-stand, buying it, and rushing home to pore over the ads, looking for a sweet new motherboard, some RAM, maybe a Sound Blaster card, etc. for that sweet new PC you were planning to build. And the BBS listings... A new BBS appearing in there, IN YOUR LOCAL CALLING AREA was a huge deal.
I know the younger folks probably don't get the idea of "long distance" or probably even the idea of phone usage being metered and billed per unit, but back in the day calling a BBS that was long-distance was a good way to rack up a huge phone bill, and righteously piss off your parents (if you lived at home). Of course some people would pack a laptop (yes, they existed in the 1990's) in a car, drive out to a convenience store in the middle of nowhere at 2:00 in the morning, and "beige box" off a COCOT (customer owned coin operated telephone) to make those calls. I might have been guilty of doing something like that...
Anyway, Computer Shopper was a big deal to a lot of people for a long time, and I'm personally thrilled to see these issues being preserved so some of that legacy can live on. Browsing the old issues online will never give one quite the same experience, but it's good that they'll be around.
I've said it before, The IA is a crown jewel of the internet. But now I'll add "and God Bless Jason Scott".
This might be one of the most important historical pricing repositories that track tech trends ever produced when it's done. Imagine tracking the weekly per-MB prices of hard drives or even more volatile DRAM from 1979 to 2009. And then there's tracking the emergence and transition between technologies, correlation against production yield information, corporate histories...this could be a backbone all other North American computer history could be hung off of.
It's a time long gone, but 2009 wasn't that long ago, and reaching from the beginning of personal computing to then is an amazing thing.
I work very long hours for the Internet Archive, and I am making a huge difference in the world working for them. It wouldn’t be right or useful for me to take on any other job. I also don’t want to be doing something like making “stuff” that I sell or otherwise speculate into some market. Leave aside I have these documentaries to finish, and time has to be short.
In a world where people are rushing to pump out AI generated content, polluting the internet, all for a buck - people like Jason also exist. What a gem of a person!
I was skimming though the pages of this issue from 1986 and thought how those prices for hardware seem fairly reasonable. But looking it up, the inflation since 1986 was ~176%, so 1 USD back then equals about 2.76 today.
This is a perfect example of how our technologies have let us down - Computer Shopper and the like (including newspapers) were useful precisely because they could be quickly browsed and flipped through.
That's not really possible with the wretched things we call "browsers" today - Heck, even with a high-bandwidth connection and a nice 4K screen to display it, I still can't nearly as readily flip through pages of the IA Computer Shopper scans as I could through the actual physical media. Browsing, along with its attendant and deliberate page-turning to move on to something new, have been replaced by endless scrolling of sameness. The Ads are all dynamic, so even though they're the things that catch your eye the most, they're not usable as navigation context markers anymore - note how different all the ads were, and how the uniqueness page layout became something a litlte bit memorable, so you could more easily flip back to re-find something you'd read a few dozen pages back.
Browsing is an excellent quick, easy way to MAKE CONNECTIONS between lots of different data coming from lots of different sources. That sort of thing created much more well-rounded, smarter people, and society is poorer for having moved to overly structured media. (And ads are media, too (heck they're nearly all of Computer Shopper) - just try going back to find an ad you saw on a site last week!) Static location context matters - humans are wired to think in terms of physical relativity, and our UIs give that up only at great cost...
I've been fully spoiled by PCPartPicker [0], which just invisibly filters out parts that aren't compatible with the components already selected for a build. The idea of having to manually keep track of every compatibility check at every component interface sounds like a path to debugging hell. On the other hand, that makes success a far stronger nerdcred signal.
The sheer variety of weird and wonderful ways things could be incompatible is really tamed by modern standards. I remember spending days as a kid working out obscure IRQ conflicts blindly flipping plastic jumpers, or editing config files.
I think what a lot of younger people also don’t realize is how gigantic a single issue of computer shopper was. They were as thick as phone books with thin weight pages. I remember multiple issues being well into the thousands of pages.
I don’t think younger people understand your description of how thick Computer Shopper was.
Because most likely they don’t know how thick is a phone book ;D
Once Computer Shopper became huge phone-book-sized things I quit buying them because I just didn't have the room or the time - fortunately our college library subscribed and I could peruse it there, but generally I'd read Byte, Creative Computing or Dr. Dobb's Journal instead. I recall when CS was kind of a newspaper prior to being a bound magazine format.
Computer Shopper, Dr. Dobb's, and Byte were all incredible and indispensable. I miss them all sorely and often wish that there was anything comparable in existence (online or not).
Whenever asked what period I (a proverbial young person, SWE) would go back to if I had a time machine, I always answer with roughly this period. To be alive during those days to see the development of the shoulders I now stand on would be such a treat, as I’m sure anecdotes like these only scratch the surface of all the stories to be told.
That said, if any of you oldheads have links to such things, do share - always curious to learn more about the beige age. Nostalgia is a hell of a drug and I’m here for that secondhand.
Eh, while the media seemed to be into "hackers" in the sense of people breaking into systems (and either messing them up or just to do it for bragging rights), that wasn't what most of us were into -- instead we were interested in writing software to get the most of the hardware we had (and used "hacker" in the old-school sense of expert coders).
Back in the 90s, my brother actually had a subscription to Computer Shopper (he had a side hustle building PCs for entertainment industry folks). I only ever bought it if I was planning on building something which was rare because computers were expensive back then. The last PC I built literally did not have two components from the same vendor. To be honest, I don’t miss those days. Say what you like about the “Apple tax,” it’s very nice to have one company who will take responsibility for things not working and my experience, at least, has been that they stand behind their products (at least as long as they’re under AppleCare).
> Say what you like about the “Apple tax,” it’s very nice to have one company who will take responsibility for things not working
Where's the fun in that? Building your own PC is a thrill - budgeting, chosing components, excitedly waiting for each to arrive or picking them from shelf, assembling everything and then pressing the power button to power everything up for the first time. Followed by your heart sinking when nothing happens "ohfuckohfuckohfuck, did I I just burn something out - there isn't any smoke, but I hope it's not the processor". Reseat everything, reconnect cables, get an some sign of life, but it's an error code - progress. Multiple attempts at fixing the error and finally get a POST screen. Apple can't sell that emotional experience or sense of ownership - self-built machines have a personality.
The extra RAM and/or cores I get from a self-build vs. equally priced Apple computer is just a bonus. Modern standards make surprise incompatibilities less likely than in the past, looking at the specs works 99.9% of the times
The excitement of those old days is still very much alive in the Raspberry Pi community. I just configured a new PiBox server to do some fairly serious work, including serving as the archive NAS for two ne startups. (It's more than capable of doing that work - it has more CPU power, more RAM, and more storage than Chevron's Crays had when I worked there in the '90s!)
Motherboard headers -> Cases have somehow not evolved what feels like at all. Myself and my brother just built from scratch within the past month and both fumbled the Power switch/Reset switch connections.
I honestly had to laugh at the lack of progress in the area.
The ads were much more than just "new version of the old thing" (motherboards, RAM, sound cards, video cards). About every issue, there would be something (or more than one) that I never even imagined you could do with a computer (or, usually, make into a computer peripheral). It was mind blowing to see, over and over, here was something completely new (or at least new to me). And there was something I'd seen before for less money, and something I'd seen before with better performance. Over and over, month after month. It was like watching the future on fast-forward.
>>” Of course some people would pack a laptop (yes, they existed in the 1990's) in a car, drive out to a convenience store in the middle of nowhere at 2:00 in the morning, and "beige box" off a COCOT (customer owned coin operated telephone) to make those calls. I might have been guilty of doing something like that...”
Ohhh man! I was just a few years too late for that but I remember reading and hearing about it. It was a dream of mine! Used to roam my neighborhood trying to beige box into junction boxes for the bell of it.
I think you mean red box. Red boxing was when you generated, or just recorded/played back the tones for different coins. The idea was to trick the phone into thinking you put in money. We only had one customer owned payphone we knew about that it worked on, but it was glorious.(even though we had noone to call)
A beige box is a telephone cable with alligator clips on the one end. You could ofcourse use it on a COCOT, but you could also use it on any telephone line.
I definitely mean "beige box." Red boxes were (mainly?) for telephone company phones that would recognize the coin tones. Those phones had a different line "class of service" at the switch end and the critical point is that red-boxing was the only way (post blue-box era anyway) to get free calls off of them. Well, at least that I knew of. But a COCOT had all the "pay phone magic" embedded in the phone itself - the line was a normal line, so if you clipped in at the demarc box, "upstream" of the actual phone, you were on a plain old phone line (with long-distance service) that you could use to your heart's content. And by the time frame I'm talking about, there weren't a lot of the red-box'able telephone company payphones around in our area.
If some COCOT's were also susceptible to red-boxing, that's something we never discovered. To be fair though, it never even occurred to me to check. I just had it burned into my brain at some point that red-boxing was for telco phones and not COCOT's.[1] In hindsight, that was quite possibly folklore that was wrong.
You could ofcourse use it on a COCOT, but you could also use it on any telephone line.
Yep. It just worked out that a convenience store in the middle of nowhere with a COCOT out in the parking lot was a good scenario for doing this kind of thing without getting caught. Or so I'm told... :-)
> But a COCOT had all the "pay phone magic" embedded in the phone itself - the line was a normal line
There is one still operating in the parking lot of a general store in a tiny, tiny town near me. I keep wanting to put an old-school modem on my laptop, paying it a visit, and dial up a BBS (a few of those still exist, too), just for old time's sake and maybe a nostalgia overdose.
by the time I learned about them, red boxes didnt work on telco payphones. we had one customer owned one in a pizza shop, it looked like an old fashioned beige desk phone, but had equipment tacked on to take money. That was the only one we could get it to work with.
In the late 1990's I was curating a project to scan in every LEGO instruction and catalog booklet. One big problem was Moiré patterns. 300dpi seemed to maximize the artifacts and higher dpi resulted in images that were too large to use on typical displays of the time (not to mention most users were on 28-56kbps modems!) I settled on recommending 100-150dpi which eliminated most of the artifacts but still gave usable image size/detail.
Then there were concerns about what LEGO (lawyers) would think about it. Because of this it originally didn't have thumbnails or any html on that part of the site. Just images in a directory structure exposed by mod_autoindex which give it a very utilitarian, non-commercial feel. Those fears were soon put to rest when I learned that LEGO customer service was referring people to the site when customers called in to request replacements for lost instructions.
When I was in school, there was a binding shop nearby which had a huge machine that looked pretty much like a guillotine, only shallower, that just sliced through thick books like butter. They stuck the book in a vice grip, turned a wheel which brought that beautiful sharp heavy blade down on the hopeless book stuck in the grip below and the paper shavings just blossomed out of the other side as the blade did its job.
One of the favorite things for us to do was to bring an old book to the shop and ask then to shave off few millimeters on each side (except the spine of course) - the frayed/yellow bits were gone and your book was new and born again :) This also was a great strategy if you wanted to reduce the dimensions of your book. Obviously only works with paperbacks and can't be repeated more than 3-4 times. I loved it then.
> In this case, I have to use a heat gun, aiming them at the glued issues of Computer Shopper, warming them up until the glue starts to become slightly liquid and then carefully pulling the pages apart from each other, placing them on a large table I’m working on. If the glue comes too close to the pages after I pull them apart, it actually sticks them back again. It’s a huge mess, and with hundreds of pages in a typical issue, hours of work.
I wonder if instead of using heat-gun to de-glue (?), it would be easier to just shave off an ever thin slice of that spine?
Hydraulic drive paper cutters are great for high-volume print finishing (I used several in print production for about two years), but I’d be wary of them for this use case:
- Waste is high. To trim off a spine, you likely have to take more of the margin than you’d like (if you trim too close to the edge of a glued spine, the risk is higher that you may bludgeon the stack of pages under pressure instead of cutting it). Not great for double page spreads or Computer Shopper’s anaemic margins.
- Failure rate is non-zero. If you’re cutting a fresh print run, no big deal, you can reprint if you screw up. If you’re cutting rare magazines, you probably don’t want to lose many to slippage or mis-cuts.
- The blade needs regular sharpening, and it’s not something you can easily do in a home workshop, or that you probably want to do (huge heavy razor blades turn out to be quite dangerous).
What _might_ help Jason more is a “thermal binding machine”, designed to bind or debind a glued spine. He could (in theory) leave each mag in the debinder for a set time to heat the glue, then pry the pages free. Not a huge improvement over a hairdryer, but at least more hands-free, especially if people could donate or loan multiple machines.
I briefly experiment with that for books before switching to a guillotine. It was slower, and made a mess. Even when I thought all the dust/grit was gone, some snuck into the scans anyway. For softer lower-quality aged acid-attacked newsprint-style magazines like this, I imagine the mess would be even worse.
I don’t know what the actual paper weight is but it looks in the “bible paper and cheap newsprint” range (~40-60gsm), which will probably rip as soon as a belt/orbital sander crosses the glue line into the looser leaves (and you have to cross the glue line to debind, and by nature that line isn’t normally straight, because glue does not melt evenly). You also destroy the spine, making it harder to identify issues at a glance.
A thermal binding machine would likely melt the glue in 45-90 seconds without those risks. And you might even be able to rebind after the scan with fresh glue strips using the same machine, assuming you don’t have to cut the cover wrap.
Using heat is common in a similar hobby(?), translating (older) Japanese manga. You would use a hair dryer or clothesiron to melt the glue in the spine, pull apart the pages/clean off the glue, and scan them in. That way you don't shave off 2-page spreads. (After scanning, you'd have to clean up the image a lot still, but oh well.) I can imagine here you also just don't want to damage the pages, even if the damage seems infinitesimal.
It doesn't happen so much anymore for manga since things are mostly digital. Though some publishers still publish at 1200x800 (looking at you, Hakusensha!) or worse (I think one site uses like, 900x600?). Mostly these days this would only happen for self-published works (which are still distributed in limited quantities and in physical format only, oftentimes). Or apparently publishers occasionally just...lose the digital files. (Though I can't find where I read this anymore.)
Also, people have already contributed thousands for this project, asking for a cutty-cutty machine is a bit much, especially when I'm really getting good with the heat gun.
Scientists routinely request more grant money for research whose difficulty exceeds initial estimates. You're already doing an incredible thing; it is okay to ask for more assistance.
If you are okay with cutting the spine off, you don’t need a big hydraulic cutter to do it (even though it is objectively the coolest way). Clamping a straightedge to the book and running a knife along that edge will cut a handful of pages very cleanly, you just run the knife through over and over. A metal ruler and a Stanley knife will do this just fine - heavy pressure on the ruler and light passes with the Stanley knife. (This is what amateur bookbinders do to trim the edges of books they’re binding when they don’t have access to proper guillotines.)
So many pages were the same unchanged ads in the Shopper every month. Can't you add in some prefiltering for those to reduce the number of pages actually needed to scan?
(Sorry for hijacking the thread...)
These will easily take a hand off so most of the machines designed after the mid-80s required the operator to press two buttons as a safety to cut and a pedal to initiate the clamp. The older ones from the 70s just required a push on a pair of foot pedals, or a pedal and a single button -- which was obviously stupid. Once the safer machines came around we got those.
I used to operate one as a kid in middle/high school before I went off to college. You can fit measure to sub-mm precision before clamping and making the cut. It was a really beautiful thing to see it cleave the edges off of several inches of paper like a hot knife cutting through room temperature butter. A real shame filling up bins and bins and bins of cuttings to toss in the trash (this was before recycling was a common thing).
Changing the blades was quite the ordeal and you had to do it frequency to send off for regrinding. At least on the machines we had. They're basically 20+lb razer blades.
Source: I grew up in a family printing business and we had machines like this to cut the margins and printer marks off of the edges of fresh printed material to help get it ready for binding or delivery. Almost every printed page goes through something like this.
Oh, those machines are pretty satisfying to use and watch. In the 00's I did a lot of IT support for print shops (I still hate the RIP server and the Fiery, although I haven't seen one for 20 years) and the guillotines fascinated me.
Considering the prices adjusted for inflation, and how little computers could do out of the box beyond the business applications they were designed for in that era, it was really quite an expensive hobby to be a hacker!
I can't even wrap my head around some of these prices... One of those pages for "February 1986" has an ad for "diskettes" box of 10 priced between $1195 and $2995. Are these what I think they are, regular floppy disks for almost $300 each?
Edit: that must have been a typo, I see prices in the same issue of about $1/disk.
I got an Atari STE in 1988, it used double-sided regular density 3.5" disks, with a capability of 720kB each, and back then it was a bargain to find no-name ones for ~$1, while the quality name-brand ones (Verbatim! Maxell!) cost about the double.
I think I’ve been paying close to the same price for computers for 25 years. Moose’s Law and relentless miniaturization and cost optimization are the only reason why.
Every time I see an archival process like this I'm amazed at all the effort and wonder if we'll ever get something like the process described in "Rainbows End" by Vernor Vinge where a library is archived by essentially shredding the books and blowing that shredded material through a tube stuffed with high speed high quality cameras and just gluing them all back together in software. It's a very fun book that cranks up to a surprisingly exciting ending.
That's essentially how DNA sequencing works. The longest human chromosome (chr1) has about 250M bases. The DNA samples are sonicated to split them into small sequences, and then sequenced en mass 100-300bps at a time. The results are then reassembled into contigs.
I thought craig ventner had a boat that did something similar to the ocean. It would scoop up genetic material on its travels, sequence what it found and store the data.
I think that's where the author pulled the idea from in the novel it's called the Librareome Project and a character brings up the similarity of the processes.
The variability from one contig build to another is surprisingly big sometimes, lending some credit to the idea that Vernon's shotgun transcriptome approach probably has multiple "optimal" solutions.
The author is both barking mad and performing a very important service. Ephemera like the Comouter Shopper will be valuable to some historians in the future, and by its very nature this is something that needs to be preserved while it’s just an amusement.
the author is also self aware (cf the Byte episode) which is unusual.
I am fascinated that the computer shopper never made the transition to the web. They could have been the ur-vertical site. This shows how pernicious path dependence is.
There was a 2 decade gap between Computer Shopper's desire to go digital (they did create a fairly useless CD version later) and the ability of networks and browsers to even poorly render digital content. Even flipping pages today at the archive.org link isn't nearly as speedy as flipping real pages...
When I was in college in 1991, I used to get Computer Shopper and spend hours scouring it for a laptop (not a common thing then). I wanted one so bad, but was sooo broke you can't even imagine. There were always random resellers getting rid of old laptops and I'd spend hours trying to figure out which one I might get. "That 286 with the flip-up half-screen black and white liquid crystal display is relatively cheap... but so clunky. Maybe I should save up for a 386? But all the affordable ones have horrible passive LCD screens that ghost. Wait... maybe if I sell a kidney..."
I never ended up ordering one, but I remember fondly the hours of dreaming about it.
I drooled over the DG/One and the Kaypro 2000 - two of the very first true "laptop" computers. (No, the GriD doesn't count - it had to be plugged in!)
Way out of my budget, so I settled for a surplus, but new Commodore SX-64 luggable, with a fancy built-in 5-inch COLOR screen! Did the usual Commodore stuff on it, then got GEOS (made a C-64 a reasonable facsimile of a Macintosh - really) and used it to write all my final college papers and even my first resume. I still have it (in its original packing) up in a closet - need to pull it down sometime and figure out why the video sync doesn't work...
I find myself doing this with modern configurators from e.g. System76 even though I have no earthly use for such an over-powered system.
The Maplin Catalogue back in the day was my choice for idle window-shopping; one could page through it for the components of some very castles in the sky imagined diy computer or gadget - none of which I could have successfully built even if I'd had the budget. Top nerd nostalgia.
I wonder if this would be useful as a dataset for economics research. Lots you could conceivably pull from the ads: fine grained pricing, company formation and destruction, trends in standards for components, performance specs of components over time. All at the time of the Cambrian explosion of computing
Also wow this is a gargantuan effort. I scanned thee equivalent of like one issue of Computer Shopper one time and it was painful. Kudos to the author.
> The presenter mentions that about 40% of books scanned have some kind of problem, either a folded or a torn page, but they’ve done some work on preventing problems.
Even if every page in 40% of books were damaged, that would only be 40% of pages, not all of them. Although your numbers may be incredibly wrong, your steel-manned assumed thesis is valid: with historical artifacts you can only accept a lower % of damaged pages than human-powered scanning would cause.
The Feb 1986 issue linked in the article caught my attention, since it is the month I was born.
Page 10 starts with a story of someone planning to build a case for the Apple I, to take it to a computer show. The guy called Steve Jobs, and Jobs offered him to send the new tape interface, which Wozniak just finished. It is an utterly amazing read!
I thought the point in debinding was to use an autofeeder - to make the actual scanning less labor intensive.
Wouldn't a camera based scanner be better? It still requires manual page turning, but will save on the prep work and destroying the magazines, even if just to sell on.
The paper and ink quality will present they're own issues, but maybe those will be quicker to solve than the debinding and page feeding process.
In magazines there's a thing called full bleed, where the image goes to the edge of the paper, and the gutter between pages is glued together so cutting, shaving or flattening will still lose some image as the binding process isn't consistent, so to save as much image as possible, they're going to the trouble to be the archival standard, after all, he decided the deglue is best. Great article.
What a wonderful resource. When I was in grad school, somebody would buy the latest Computer Shopper and it would be passed around as needed. You really didn't need a brand new copy, so the old ones were passed around too. Sometimes it just gave you an idea of what could be built, and what kind of planning was needed. And payment / shipment were nontrivial in those days.
Info didn't age as fast then. The same products were being sold, and for about the same price, for MANY months. And there just weren't very many products. A full-page ad in a pretty large font could list every product that existed in the consumer space.
Library/archive/historical-preservation people these days, for a while now, generally don't use "scanners" to digitize historical printed/2D material images, but ordinary high-quality digital cameras, set on tripods, often downward-facing, and manual page-turning.
(although that site seems to be unmaintained with many broken links, it has enough working links you should be able to get the idea. This was a much more popular hobbyist DIY thing a decade+ ago, in the era of dawn of google books etc).
I wonder if there were particular reasons OP went with a "scanner". it seems a lot of the challenge was finding an affordable scanner that could handle the material. Perhaps it ends up being much (much) faster per-page, especially if you are willing to destroy the original, and is so justified? (Even though they ended up having to do manual page feed, possibly still much faster?) Or if they just didn't think of camera-tripod-manually-turned-pages and know it was a (common) option? If it hasn't been considered, it might be worth considering, although the process they have seems to be working well -- the sample output is stunning!
There are a number of commercial substantially sub-$1k prosumer overhead book scanner options now that have nicer work-flows and more automated post processing than the DIY book scanners folks used to hack up (or hand feeding a flatbed) - I think they came out of and subsequently took took a lot of the air out of the DIY efforts. CZUR has been pretty generous with giving book scanning rigs ( https://www.czur.com/product/ETscanner ) to prominent hobbyists (I think CuriousMarc and Shelby from TechTangents and a couple other retrocomputing type youtube folks got them - apparently it was a good marketing move, that's why I'm aware of them).
The two downsides to those rigs here are that it looks like this effort is aiming for 600dpi capture and that would be a ...reach... for the reasonable cost overhead systems, and (Using it as an example because I've looked at theirs) CZUR's machines only go up to about an A3 scan area, while computer shopper was a ridiculous 13x10" format which would be a bit too big to do in the expected 2-up workflow with their units.
I feel like 600dpi ought to be possible with a DIY setup and a good quality but not break-the-bank camera, but maybe I'm wrong?
Where I work, I am not the one who works with the photographic digitization. I know we can do 600 dpi digitization of things of this size (whether we choose to depends on the material). But I suppose our setup may cost substantially more than OP, I'm not sure, I could ask the relevant staff.
Sorry, I did not mean to be trying to get in OP's business, who I assume you are. I was just curious to learn more about comparison between your choice and non-"scanner" suspended camera type rigs which I don't believe your interesting piece covered, so was interested in working that through with other commenters.
This is something I know a little about. Back in 2005/06 I asked Google who was doing non-destructive scanning of their books as I needed to scan our back issues to 1948 and no one would let me cut up bound volumes. They put me in touch with Kirtas Technologies that did the job for something like $20K and took two weeks (about 70 volumes). It's a lot cheaper now and the scanning quality is significantly better but the big headache is indexing.
Since then I've been collecting spare issues of the magazine and last year sent a decade's worth off to be destructively scanned for about $6K. The quality is significantly better than the originals (and they have to do clean up as part of the process).
I point this out to show what a challenging job is being attempted here and how really there is value and hard work in digitizing archives so there really shouldn't be an expectation of old stuff being 'free'.
Sometime I feel sad that younger generations won’t have the excitements of early days of PC and sometimes I feel they might be better off spending their time on something that is more productive.
One thing for sure, that era is gone and all
The most of the enablers of that era are also history now.
Amount of time I spend reading and researching which motherboard and RAM would best fit together makes me smile from time to time. Or the amount of time I spend searching for extra memory chips for my trident card…
(Also: completely agree - this stuff is going to be a fabulous resource for future researchers and meanwhile the rest of us reap the benefit by indulging our nostalgia - I'm full of admiration)
I used to buy the British edition at least partially for the articles, but it could be a damned hard slog finding them.
In fact a lot of the computer magazines tended in this direction - presumably the consolidation around the PC and the commoditization of it ate into margins as they all focused on the one (albeit growing) narrower topic.
By the by, I often joke that I buy The Economist for the pictures (not true, but the captions are often very sly jokes).
That must be the Holy Grail of publishing. You're buying it for the ads. So nobody complains that it's packed absolutely full of ads. And they're selling it to you. That's amazing. (And it was true! We did want to see the ads!)
I think I got my 1st custom PC from a vendor in CS. It was a Pentium 1 166Mhz MMX / 8MB RAM and 1.2GB HDD.
This is the box that I wrote a lot of VB6 code on for AOL, pirated my 1st MP3 (Deftones My Own Summer), and found my favorite SNES game (FF4). I didn't even know emulators were a thing!
With windows 98 and with 8MB of ram, i really learned about startup items.
There was a GE capital nearby that had a dumpster full of computer parts. I ended up finding some 8MB EDO sticks in the dumpster and it was the 1st time my computer would idle without trashing the HDD. 32MB was epic.
> The cheap, cheap paper is a nightmare to run through a scanner
Not to mention having a faint mirror image of the other side. You need a dark background and a lot of light on the front to catch as little of it as possible.
And, then, one could add some smarts to try to align and subtract the image of the other side of the page to get an even cleaner image.
Now, just imagine if they kept an archive of their photo plates (or, better for later issues, PDF copies). By 2009 that's the very least one should have.
The regulars, including Don Lancaster, were great. I would drool over so many articles that I barely understood, and one of my goals for the next month were to get caught-up on some part of this tech that had escaped me.
Computer Shopper was fantastic for guiding me to complement my more formal studies.
OT: Does anyone else still use 6502/65c816's and TMS32C32's?
I seriously thought about bidding on that ebay auction when it came up - I'm glad it went to a far better place! Good luck to him indeed - growing up in that time Computer Shopper was the logical progression from the Sears Wishbook.
Apart from nostalgia, it contains a staggering amount of information about historical PC pricing and hardware availability that (to my knowledge) isn't really documented anywhere else. Sure, Wikipedia can tell you the official launch date of the Intel Pentium, but Computer Shopper can tell you when the price actually dropped to something reasonable, when cheap motherboards for it started pouring out of no-name manufacturers, when vendors stopped bothering to advertise 486 hardware, etc..
This... is fugging amazing.
No, seriously. It really is. Those of you who are too young to remember the heyday of Computer Shopper won't get it (probably) but that magazine was ... well, as the article says, practically a "bible" to some of us. There was NOTHING like finding the new issue on the news-stand, buying it, and rushing home to pore over the ads, looking for a sweet new motherboard, some RAM, maybe a Sound Blaster card, etc. for that sweet new PC you were planning to build. And the BBS listings... A new BBS appearing in there, IN YOUR LOCAL CALLING AREA was a huge deal.
I know the younger folks probably don't get the idea of "long distance" or probably even the idea of phone usage being metered and billed per unit, but back in the day calling a BBS that was long-distance was a good way to rack up a huge phone bill, and righteously piss off your parents (if you lived at home). Of course some people would pack a laptop (yes, they existed in the 1990's) in a car, drive out to a convenience store in the middle of nowhere at 2:00 in the morning, and "beige box" off a COCOT (customer owned coin operated telephone) to make those calls. I might have been guilty of doing something like that...
Anyway, Computer Shopper was a big deal to a lot of people for a long time, and I'm personally thrilled to see these issues being preserved so some of that legacy can live on. Browsing the old issues online will never give one quite the same experience, but it's good that they'll be around.
I've said it before, The IA is a crown jewel of the internet. But now I'll add "and God Bless Jason Scott".
This might be one of the most important historical pricing repositories that track tech trends ever produced when it's done. Imagine tracking the weekly per-MB prices of hard drives or even more volatile DRAM from 1979 to 2009. And then there's tracking the emergence and transition between technologies, correlation against production yield information, corporate histories...this could be a backbone all other North American computer history could be hung off of.
It's a time long gone, but 2009 wasn't that long ago, and reaching from the beginning of personal computing to then is an amazing thing.
edit: wow, and I mean wow the results look great https://archive.org/details/computer-shopper-february-1986/C...
edit2: WHAT, THEY'RE ALSO OCR'd???!?!?! Holy Fuck
edit3: if you don't, you should listen to Jason's podcast, it really helps get into the mind of an amazing human https://www.youtube.com/@jasonscott526
I work very long hours for the Internet Archive, and I am making a huge difference in the world working for them. It wouldn’t be right or useful for me to take on any other job. I also don’t want to be doing something like making “stuff” that I sell or otherwise speculate into some market. Leave aside I have these documentaries to finish, and time has to be short.
from http://ascii.textfiles.com/archives/5251
In a world where people are rushing to pump out AI generated content, polluting the internet, all for a buck - people like Jason also exist. What a gem of a person!
Saint Jason Scott of the Internet is how the machines will remember him.
I was skimming though the pages of this issue from 1986 and thought how those prices for hardware seem fairly reasonable. But looking it up, the inflation since 1986 was ~176%, so 1 USD back then equals about 2.76 today.
finger guns
You can just smell the pages as you flip through.
This is a perfect example of how our technologies have let us down - Computer Shopper and the like (including newspapers) were useful precisely because they could be quickly browsed and flipped through.
That's not really possible with the wretched things we call "browsers" today - Heck, even with a high-bandwidth connection and a nice 4K screen to display it, I still can't nearly as readily flip through pages of the IA Computer Shopper scans as I could through the actual physical media. Browsing, along with its attendant and deliberate page-turning to move on to something new, have been replaced by endless scrolling of sameness. The Ads are all dynamic, so even though they're the things that catch your eye the most, they're not usable as navigation context markers anymore - note how different all the ads were, and how the uniqueness page layout became something a litlte bit memorable, so you could more easily flip back to re-find something you'd read a few dozen pages back.
Browsing is an excellent quick, easy way to MAKE CONNECTIONS between lots of different data coming from lots of different sources. That sort of thing created much more well-rounded, smarter people, and society is poorer for having moved to overly structured media. (And ads are media, too (heck they're nearly all of Computer Shopper) - just try going back to find an ad you saw on a site last week!) Static location context matters - humans are wired to think in terms of physical relativity, and our UIs give that up only at great cost...
[dead]
I've been fully spoiled by PCPartPicker [0], which just invisibly filters out parts that aren't compatible with the components already selected for a build. The idea of having to manually keep track of every compatibility check at every component interface sounds like a path to debugging hell. On the other hand, that makes success a far stronger nerdcred signal.
[0] https://pcpartpicker.com/list/
The sheer variety of weird and wonderful ways things could be incompatible is really tamed by modern standards. I remember spending days as a kid working out obscure IRQ conflicts blindly flipping plastic jumpers, or editing config files.
I don't miss it!
My first use of spreadsheets was to track IRQ, IO Port and memory address conflicts :p
I remember when this was just something the creator was messing with. Its come a long way.
Parts compatability isn't really too hard. Pull up the mobo specs and make sure your ram timings, CPU socket, PCI, and power demand all line up.
I think what a lot of younger people also don’t realize is how gigantic a single issue of computer shopper was. They were as thick as phone books with thin weight pages. I remember multiple issues being well into the thousands of pages.
I don’t think younger people understand your description of how thick Computer Shopper was. Because most likely they don’t know how thick is a phone book ;D
For those that don't know, a phone book was often about twice as thick as a Thomas Guide.
That should help.
And a Thomas Guide was about as thick as 8 or 9 J.C. Whitney catalogs.
Another set of imperial units unknown to us Europeans? :-) How tick was that, a couple of inches / 5 cm?
1/2000 of a football field, give or take
American Football, or soccer?
It would be about 6 American Football compact disk jewel cases.
American of course, the other sort isn't a unit of measurement!
That couldn't even fit on a floppy disk. You know, that icon thing kids click on to save things.
15-20 Farmer’s Almanacs.
So around 6 iphones?
Once Computer Shopper became huge phone-book-sized things I quit buying them because I just didn't have the room or the time - fortunately our college library subscribed and I could peruse it there, but generally I'd read Byte, Creative Computing or Dr. Dobb's Journal instead. I recall when CS was kind of a newspaper prior to being a bound magazine format.
Computer Shopper, Dr. Dobb's, and Byte were all incredible and indispensable. I miss them all sorely and often wish that there was anything comparable in existence (online or not).
I used to use old ones as rifle target backstops. .22LR would not penetrate a single issue.
It's very brave to mention such heinous crimes in this discussion.
I imagine that at the time they were just as ubiquitous as the phone books.
Fair. I guess historians would also object against destroying phone books, if not right now, in a hundred years.
We recently told our daughter what a phone book was. She was flabbergasted.
I should brag on my 1988 - 1996 complete run of "PC Magazine" too
Whenever asked what period I (a proverbial young person, SWE) would go back to if I had a time machine, I always answer with roughly this period. To be alive during those days to see the development of the shoulders I now stand on would be such a treat, as I’m sure anecdotes like these only scratch the surface of all the stories to be told.
That said, if any of you oldheads have links to such things, do share - always curious to learn more about the beige age. Nostalgia is a hell of a drug and I’m here for that secondhand.
I think this book captures a good bit of the spirit
https://www.goodreads.com/book/show/984598.Masters_of_Decept...
Eh, while the media seemed to be into "hackers" in the sense of people breaking into systems (and either messing them up or just to do it for bragging rights), that wasn't what most of us were into -- instead we were interested in writing software to get the most of the hardware we had (and used "hacker" in the old-school sense of expert coders).
Steven levy, “hackers: heroes of the computer revolution” probably a bit more your tune, maybe.
Yes, I've read that.
[dead]
Back in the 90s, my brother actually had a subscription to Computer Shopper (he had a side hustle building PCs for entertainment industry folks). I only ever bought it if I was planning on building something which was rare because computers were expensive back then. The last PC I built literally did not have two components from the same vendor. To be honest, I don’t miss those days. Say what you like about the “Apple tax,” it’s very nice to have one company who will take responsibility for things not working and my experience, at least, has been that they stand behind their products (at least as long as they’re under AppleCare).
> Say what you like about the “Apple tax,” it’s very nice to have one company who will take responsibility for things not working
Where's the fun in that? Building your own PC is a thrill - budgeting, chosing components, excitedly waiting for each to arrive or picking them from shelf, assembling everything and then pressing the power button to power everything up for the first time. Followed by your heart sinking when nothing happens "ohfuckohfuckohfuck, did I I just burn something out - there isn't any smoke, but I hope it's not the processor". Reseat everything, reconnect cables, get an some sign of life, but it's an error code - progress. Multiple attempts at fixing the error and finally get a POST screen. Apple can't sell that emotional experience or sense of ownership - self-built machines have a personality.
The extra RAM and/or cores I get from a self-build vs. equally priced Apple computer is just a bonus. Modern standards make surprise incompatibilities less likely than in the past, looking at the specs works 99.9% of the times
The excitement of those old days is still very much alive in the Raspberry Pi community. I just configured a new PiBox server to do some fairly serious work, including serving as the archive NAS for two ne startups. (It's more than capable of doing that work - it has more CPU power, more RAM, and more storage than Chevron's Crays had when I worked there in the '90s!)
Motherboard headers -> Cases have somehow not evolved what feels like at all. Myself and my brother just built from scratch within the past month and both fumbled the Power switch/Reset switch connections.
I honestly had to laugh at the lack of progress in the area.
The ads were much more than just "new version of the old thing" (motherboards, RAM, sound cards, video cards). About every issue, there would be something (or more than one) that I never even imagined you could do with a computer (or, usually, make into a computer peripheral). It was mind blowing to see, over and over, here was something completely new (or at least new to me). And there was something I'd seen before for less money, and something I'd seen before with better performance. Over and over, month after month. It was like watching the future on fast-forward.
>>” Of course some people would pack a laptop (yes, they existed in the 1990's) in a car, drive out to a convenience store in the middle of nowhere at 2:00 in the morning, and "beige box" off a COCOT (customer owned coin operated telephone) to make those calls. I might have been guilty of doing something like that...”
Ohhh man! I was just a few years too late for that but I remember reading and hearing about it. It was a dream of mine! Used to roam my neighborhood trying to beige box into junction boxes for the bell of it.
>Used to roam my neighborhood trying to beige box into junction boxes for the bell of it.
I'd be very surprised if this wasn't accidental given the character of this site, but holy moly is the pun fitting. Congrats.
...and "beige box" off a COCOT...
I think you mean red box. Red boxing was when you generated, or just recorded/played back the tones for different coins. The idea was to trick the phone into thinking you put in money. We only had one customer owned payphone we knew about that it worked on, but it was glorious.(even though we had noone to call)
A beige box is a telephone cable with alligator clips on the one end. You could ofcourse use it on a COCOT, but you could also use it on any telephone line.
I definitely mean "beige box." Red boxes were (mainly?) for telephone company phones that would recognize the coin tones. Those phones had a different line "class of service" at the switch end and the critical point is that red-boxing was the only way (post blue-box era anyway) to get free calls off of them. Well, at least that I knew of. But a COCOT had all the "pay phone magic" embedded in the phone itself - the line was a normal line, so if you clipped in at the demarc box, "upstream" of the actual phone, you were on a plain old phone line (with long-distance service) that you could use to your heart's content. And by the time frame I'm talking about, there weren't a lot of the red-box'able telephone company payphones around in our area.
If some COCOT's were also susceptible to red-boxing, that's something we never discovered. To be fair though, it never even occurred to me to check. I just had it burned into my brain at some point that red-boxing was for telco phones and not COCOT's.[1] In hindsight, that was quite possibly folklore that was wrong.
You could ofcourse use it on a COCOT, but you could also use it on any telephone line.
Yep. It just worked out that a convenience store in the middle of nowhere with a COCOT out in the parking lot was a good scenario for doing this kind of thing without getting caught. Or so I'm told... :-)
[1]: https://www.tech-faq.com/red-box.html
> But a COCOT had all the "pay phone magic" embedded in the phone itself - the line was a normal line
There is one still operating in the parking lot of a general store in a tiny, tiny town near me. I keep wanting to put an old-school modem on my laptop, paying it a visit, and dial up a BBS (a few of those still exist, too), just for old time's sake and maybe a nostalgia overdose.
by the time I learned about them, red boxes didnt work on telco payphones. we had one customer owned one in a pizza shop, it looked like an old fashioned beige desk phone, but had equipment tacked on to take money. That was the only one we could get it to work with.
It’s kind of amazing to think of how profoundly boring my life might have been if there was no such thing as long distance charges.
It was the one thing I carried everywhere as a teenager.
Yep, been there, done that.
[flagged]
[flagged]
[flagged]
In the late 1990's I was curating a project to scan in every LEGO instruction and catalog booklet. One big problem was Moiré patterns. 300dpi seemed to maximize the artifacts and higher dpi resulted in images that were too large to use on typical displays of the time (not to mention most users were on 28-56kbps modems!) I settled on recommending 100-150dpi which eliminated most of the artifacts but still gave usable image size/detail.
Then there were concerns about what LEGO (lawyers) would think about it. Because of this it originally didn't have thumbnails or any html on that part of the site. Just images in a directory structure exposed by mod_autoindex which give it a very utilitarian, non-commercial feel. Those fears were soon put to rest when I learned that LEGO customer service was referring people to the site when customers called in to request replacements for lost instructions.
That archive lives on today at http://www.peeron.com/ (not my site)
> debinding
When I was in school, there was a binding shop nearby which had a huge machine that looked pretty much like a guillotine, only shallower, that just sliced through thick books like butter. They stuck the book in a vice grip, turned a wheel which brought that beautiful sharp heavy blade down on the hopeless book stuck in the grip below and the paper shavings just blossomed out of the other side as the blade did its job.
One of the favorite things for us to do was to bring an old book to the shop and ask then to shave off few millimeters on each side (except the spine of course) - the frayed/yellow bits were gone and your book was new and born again :) This also was a great strategy if you wanted to reduce the dimensions of your book. Obviously only works with paperbacks and can't be repeated more than 3-4 times. I loved it then.
> In this case, I have to use a heat gun, aiming them at the glued issues of Computer Shopper, warming them up until the glue starts to become slightly liquid and then carefully pulling the pages apart from each other, placing them on a large table I’m working on. If the glue comes too close to the pages after I pull them apart, it actually sticks them back again. It’s a huge mess, and with hundreds of pages in a typical issue, hours of work.
I wonder if instead of using heat-gun to de-glue (?), it would be easier to just shave off an ever thin slice of that spine?
(sorry if that was too gruesome for you).
Hydraulic drive paper cutters are great for high-volume print finishing (I used several in print production for about two years), but I’d be wary of them for this use case:
- Waste is high. To trim off a spine, you likely have to take more of the margin than you’d like (if you trim too close to the edge of a glued spine, the risk is higher that you may bludgeon the stack of pages under pressure instead of cutting it). Not great for double page spreads or Computer Shopper’s anaemic margins.
- Failure rate is non-zero. If you’re cutting a fresh print run, no big deal, you can reprint if you screw up. If you’re cutting rare magazines, you probably don’t want to lose many to slippage or mis-cuts.
- The blade needs regular sharpening, and it’s not something you can easily do in a home workshop, or that you probably want to do (huge heavy razor blades turn out to be quite dangerous).
What _might_ help Jason more is a “thermal binding machine”, designed to bind or debind a glued spine. He could (in theory) leave each mag in the debinder for a set time to heat the glue, then pry the pages free. Not a huge improvement over a hairdryer, but at least more hands-free, especially if people could donate or loan multiple machines.
What if instead of cutting, we sanded them? The pressure on the surfaces would be smaller and it could remove just enough to debind the volume.
I briefly experiment with that for books before switching to a guillotine. It was slower, and made a mess. Even when I thought all the dust/grit was gone, some snuck into the scans anyway. For softer lower-quality aged acid-attacked newsprint-style magazines like this, I imagine the mess would be even worse.
Nice idea but I would be worried about tearing.
I don’t know what the actual paper weight is but it looks in the “bible paper and cheap newsprint” range (~40-60gsm), which will probably rip as soon as a belt/orbital sander crosses the glue line into the looser leaves (and you have to cross the glue line to debind, and by nature that line isn’t normally straight, because glue does not melt evenly). You also destroy the spine, making it harder to identify issues at a glance.
A thermal binding machine would likely melt the glue in 45-90 seconds without those risks. And you might even be able to rebind after the scan with fresh glue strips using the same machine, assuming you don’t have to cut the cover wrap.
Using heat is common in a similar hobby(?), translating (older) Japanese manga. You would use a hair dryer or clothesiron to melt the glue in the spine, pull apart the pages/clean off the glue, and scan them in. That way you don't shave off 2-page spreads. (After scanning, you'd have to clean up the image a lot still, but oh well.) I can imagine here you also just don't want to damage the pages, even if the damage seems infinitesimal.
It doesn't happen so much anymore for manga since things are mostly digital. Though some publishers still publish at 1200x800 (looking at you, Hakusensha!) or worse (I think one site uses like, 900x600?). Mostly these days this would only happen for self-published works (which are still distributed in limited quantities and in physical format only, oftentimes). Or apparently publishers occasionally just...lose the digital files. (Though I can't find where I read this anymore.)
> That way you don't shave off 2-page spreads.
A very good point
Also, people have already contributed thousands for this project, asking for a cutty-cutty machine is a bit much, especially when I'm really getting good with the heat gun.
Scientists routinely request more grant money for research whose difficulty exceeds initial estimates. You're already doing an incredible thing; it is okay to ask for more assistance.
If you are okay with cutting the spine off, you don’t need a big hydraulic cutter to do it (even though it is objectively the coolest way). Clamping a straightedge to the book and running a knife along that edge will cut a handful of pages very cleanly, you just run the knife through over and over. A metal ruler and a Stanley knife will do this just fine - heavy pressure on the ruler and light passes with the Stanley knife. (This is what amateur bookbinders do to trim the edges of books they’re binding when they don’t have access to proper guillotines.)
So many pages were the same unchanged ads in the Shopper every month. Can't you add in some prefiltering for those to reduce the number of pages actually needed to scan? (Sorry for hijacking the thread...)
No
Upvoted with extra strong mouse click on account of cutty-cutty machine.
Something like this? https://www.whitakerbrothers.com/products/martin-yale-7000e
Probably more like one of these, a hydraulic guillotine cutter.
https://youtu.be/RZD0XHp75S4
https://youtu.be/BvLk6-OugIc
These will easily take a hand off so most of the machines designed after the mid-80s required the operator to press two buttons as a safety to cut and a pedal to initiate the clamp. The older ones from the 70s just required a push on a pair of foot pedals, or a pedal and a single button -- which was obviously stupid. Once the safer machines came around we got those.
I used to operate one as a kid in middle/high school before I went off to college. You can fit measure to sub-mm precision before clamping and making the cut. It was a really beautiful thing to see it cleave the edges off of several inches of paper like a hot knife cutting through room temperature butter. A real shame filling up bins and bins and bins of cuttings to toss in the trash (this was before recycling was a common thing).
Changing the blades was quite the ordeal and you had to do it frequency to send off for regrinding. At least on the machines we had. They're basically 20+lb razer blades.
Source: I grew up in a family printing business and we had machines like this to cut the margins and printer marks off of the edges of fresh printed material to help get it ready for binding or delivery. Almost every printed page goes through something like this.
Oh, those machines are pretty satisfying to use and watch. In the 00's I did a lot of IT support for print shops (I still hate the RIP server and the Fiery, although I haven't seen one for 20 years) and the guillotines fascinated me.
Similar but 4-5 times as bigger - it could have chopped my 12 year old body in half. A beast of a machine.
(I don't know what has come over me - my mind is only producing grisly descriptions)
> huge machine that looked pretty much like a guillotine
My local binding shop actually still operates one of those.
Considering the prices adjusted for inflation, and how little computers could do out of the box beyond the business applications they were designed for in that era, it was really quite an expensive hobby to be a hacker!
I can't even wrap my head around some of these prices... One of those pages for "February 1986" has an ad for "diskettes" box of 10 priced between $1195 and $2995. Are these what I think they are, regular floppy disks for almost $300 each?
Edit: that must have been a typo, I see prices in the same issue of about $1/disk.
I saw that, too. They are missing a decimal point. 11.95 and 29.95.
I got an Atari STE in 1988, it used double-sided regular density 3.5" disks, with a capability of 720kB each, and back then it was a bargain to find no-name ones for ~$1, while the quality name-brand ones (Verbatim! Maxell!) cost about the double.
IIRC $300 each was 1 GB zip disks in the middle 1990's.
I think I’ve been paying close to the same price for computers for 25 years. Moose’s Law and relentless miniaturization and cost optimization are the only reason why.
Every time I see an archival process like this I'm amazed at all the effort and wonder if we'll ever get something like the process described in "Rainbows End" by Vernor Vinge where a library is archived by essentially shredding the books and blowing that shredded material through a tube stuffed with high speed high quality cameras and just gluing them all back together in software. It's a very fun book that cranks up to a surprisingly exciting ending.
That's essentially how DNA sequencing works. The longest human chromosome (chr1) has about 250M bases. The DNA samples are sonicated to split them into small sequences, and then sequenced en mass 100-300bps at a time. The results are then reassembled into contigs.
https://en.wikipedia.org/wiki/Chromosome_1
https://en.wikipedia.org/wiki/DNA_sequencing#High-throughput...
I thought craig ventner had a boat that did something similar to the ocean. It would scoop up genetic material on its travels, sequence what it found and store the data.
https://en.wikipedia.org/wiki/Global_Ocean_Sampling_Expediti...
(the page doesn't seem to explain it)
True, but you wouldn't do that to the very last one chromosome in existence.
Some of these tomes might well be the last copy out there.
I think that's where the author pulled the idea from in the novel it's called the Librareome Project and a character brings up the similarity of the processes.
The variability from one contig build to another is surprisingly big sometimes, lending some credit to the idea that Vernon's shotgun transcriptome approach probably has multiple "optimal" solutions.
This is how the human visual system works.
https://web.archive.org/web/20230606004015/http://ascii.text...
The author is both barking mad and performing a very important service. Ephemera like the Comouter Shopper will be valuable to some historians in the future, and by its very nature this is something that needs to be preserved while it’s just an amusement.
the author is also self aware (cf the Byte episode) which is unusual.
I am fascinated that the computer shopper never made the transition to the web. They could have been the ur-vertical site. This shows how pernicious path dependence is.
There was a 2 decade gap between Computer Shopper's desire to go digital (they did create a fairly useless CD version later) and the ability of networks and browsers to even poorly render digital content. Even flipping pages today at the archive.org link isn't nearly as speedy as flipping real pages...
eBay pulled it off. Just a search engine that pulled up a bunch of ads would have been great.
When I was in college in 1991, I used to get Computer Shopper and spend hours scouring it for a laptop (not a common thing then). I wanted one so bad, but was sooo broke you can't even imagine. There were always random resellers getting rid of old laptops and I'd spend hours trying to figure out which one I might get. "That 286 with the flip-up half-screen black and white liquid crystal display is relatively cheap... but so clunky. Maybe I should save up for a 386? But all the affordable ones have horrible passive LCD screens that ghost. Wait... maybe if I sell a kidney..."
I never ended up ordering one, but I remember fondly the hours of dreaming about it.
I drooled over the DG/One and the Kaypro 2000 - two of the very first true "laptop" computers. (No, the GriD doesn't count - it had to be plugged in!)
Way out of my budget, so I settled for a surplus, but new Commodore SX-64 luggable, with a fancy built-in 5-inch COLOR screen! Did the usual Commodore stuff on it, then got GEOS (made a C-64 a reasonable facsimile of a Macintosh - really) and used it to write all my final college papers and even my first resume. I still have it (in its original packing) up in a closet - need to pull it down sometime and figure out why the video sync doesn't work...
I find myself doing this with modern configurators from e.g. System76 even though I have no earthly use for such an over-powered system.
The Maplin Catalogue back in the day was my choice for idle window-shopping; one could page through it for the components of some very castles in the sky imagined diy computer or gadget - none of which I could have successfully built even if I'd had the budget. Top nerd nostalgia.
I wonder if this would be useful as a dataset for economics research. Lots you could conceivably pull from the ads: fine grained pricing, company formation and destruction, trends in standards for components, performance specs of components over time. All at the time of the Cambrian explosion of computing
Also wow this is a gargantuan effort. I scanned thee equivalent of like one issue of Computer Shopper one time and it was painful. Kudos to the author.
For any moderately large scale book scanning effort: https://hackaday.com/2012/11/16/google-books-team-open-sourc...
That prototype "only" damages 70% of the pages.
Really? A comment on the page I linked said:
> The presenter mentions that about 40% of books scanned have some kind of problem, either a folded or a torn page, but they’ve done some work on preventing problems.
Even if every page in 40% of books were damaged, that would only be 40% of pages, not all of them. Although your numbers may be incredibly wrong, your steel-manned assumed thesis is valid: with historical artifacts you can only accept a lower % of damaged pages than human-powered scanning would cause.
I assure you, I hate that thing. It has dogged me and the Internet Archive for over a decade. It is not just vaporware, but dangerous.
It must be a Soviet machine for scanning 40% of the pages then.
The Feb 1986 issue linked in the article caught my attention, since it is the month I was born.
Page 10 starts with a story of someone planning to build a case for the Apple I, to take it to a computer show. The guy called Steve Jobs, and Jobs offered him to send the new tape interface, which Wozniak just finished. It is an utterly amazing read!
https://archive.org/details/computer-shopper-february-1986/C...
I thought the point in debinding was to use an autofeeder - to make the actual scanning less labor intensive.
Wouldn't a camera based scanner be better? It still requires manual page turning, but will save on the prep work and destroying the magazines, even if just to sell on.
The paper and ink quality will present they're own issues, but maybe those will be quicker to solve than the debinding and page feeding process.
https://archive.org/details/tabletopscribesystem
https://twitter.com/internetarchive/status/13580909821897195...
https://blog.archive.org/2021/02/09/meet-eliza-zhang-book-sc...
https://www.diybookscanner.org/
The effort here to preserve history though is incredible.
With very thick books like those, debinding is necessary even just to get a good straight scan.
Well, there is no situation where this isn't in some way labor-intensive.
But one of the reasons that I chose to use the feed scanner that I do is that I get 600 dots per inch on these scans of these massive tabloids.
In magazines there's a thing called full bleed, where the image goes to the edge of the paper, and the gutter between pages is glued together so cutting, shaving or flattening will still lose some image as the binding process isn't consistent, so to save as much image as possible, they're going to the trouble to be the archival standard, after all, he decided the deglue is best. Great article.
Here's the collection so far: https://archive.org/search?query=creator%3A%22Computer+Shopp...
What a wonderful resource. When I was in grad school, somebody would buy the latest Computer Shopper and it would be passed around as needed. You really didn't need a brand new copy, so the old ones were passed around too. Sometimes it just gave you an idea of what could be built, and what kind of planning was needed. And payment / shipment were nontrivial in those days.
Info didn't age as fast then. The same products were being sold, and for about the same price, for MANY months. And there just weren't very many products. A full-page ad in a pretty large font could list every product that existed in the consumer space.
This is an awesome project!
Library/archive/historical-preservation people these days, for a while now, generally don't use "scanners" to digitize historical printed/2D material images, but ordinary high-quality digital cameras, set on tripods, often downward-facing, and manual page-turning.
There is various plans for DIY rigs for such on the internet, for instance here: https://www.diybookscanner.org/.
(although that site seems to be unmaintained with many broken links, it has enough working links you should be able to get the idea. This was a much more popular hobbyist DIY thing a decade+ ago, in the era of dawn of google books etc).
I wonder if there were particular reasons OP went with a "scanner". it seems a lot of the challenge was finding an affordable scanner that could handle the material. Perhaps it ends up being much (much) faster per-page, especially if you are willing to destroy the original, and is so justified? (Even though they ended up having to do manual page feed, possibly still much faster?) Or if they just didn't think of camera-tripod-manually-turned-pages and know it was a (common) option? If it hasn't been considered, it might be worth considering, although the process they have seems to be working well -- the sample output is stunning!
There are a number of commercial substantially sub-$1k prosumer overhead book scanner options now that have nicer work-flows and more automated post processing than the DIY book scanners folks used to hack up (or hand feeding a flatbed) - I think they came out of and subsequently took took a lot of the air out of the DIY efforts. CZUR has been pretty generous with giving book scanning rigs ( https://www.czur.com/product/ETscanner ) to prominent hobbyists (I think CuriousMarc and Shelby from TechTangents and a couple other retrocomputing type youtube folks got them - apparently it was a good marketing move, that's why I'm aware of them).
The two downsides to those rigs here are that it looks like this effort is aiming for 600dpi capture and that would be a ...reach... for the reasonable cost overhead systems, and (Using it as an example because I've looked at theirs) CZUR's machines only go up to about an A3 scan area, while computer shopper was a ridiculous 13x10" format which would be a bit too big to do in the expected 2-up workflow with their units.
Nice, thanks for the info.
I feel like 600dpi ought to be possible with a DIY setup and a good quality but not break-the-bank camera, but maybe I'm wrong?
Where I work, I am not the one who works with the photographic digitization. I know we can do 600 dpi digitization of things of this size (whether we choose to depends on the material). But I suppose our setup may cost substantially more than OP, I'm not sure, I could ask the relevant staff.
I gave everything a nice long think and I don't need another consultant, thanks.
Sorry, I did not mean to be trying to get in OP's business, who I assume you are. I was just curious to learn more about comparison between your choice and non-"scanner" suspended camera type rigs which I don't believe your interesting piece covered, so was interested in working that through with other commenters.
It's a great project, thanks!
This is something I know a little about. Back in 2005/06 I asked Google who was doing non-destructive scanning of their books as I needed to scan our back issues to 1948 and no one would let me cut up bound volumes. They put me in touch with Kirtas Technologies that did the job for something like $20K and took two weeks (about 70 volumes). It's a lot cheaper now and the scanning quality is significantly better but the big headache is indexing.
Since then I've been collecting spare issues of the magazine and last year sent a decade's worth off to be destructively scanned for about $6K. The quality is significantly better than the originals (and they have to do clean up as part of the process).
I point this out to show what a challenging job is being attempted here and how really there is value and hard work in digitizing archives so there really shouldn't be an expectation of old stuff being 'free'.
Sometime I feel sad that younger generations won’t have the excitements of early days of PC and sometimes I feel they might be better off spending their time on something that is more productive. One thing for sure, that era is gone and all The most of the enablers of that era are also history now.
Amount of time I spend reading and researching which motherboard and RAM would best fit together makes me smile from time to time. Or the amount of time I spend searching for extra memory chips for my trident card…
1983: you can use it as a word processor
2023: give it a prompt and it will write it on demand
2063: welp there goes the Solar System
ah the hours spent building the perfect computer on paper using arbitrarily decided limit of money I was never going to have.
There needs to be a nerdy online official saint or knighthood title for people like this.
Sounds like a job for... https://en.m.wikipedia.org/wiki/Knights_of_the_Lambda_Calcul...
(Also: completely agree - this stuff is going to be a fabulous resource for future researchers and meanwhile the rest of us reap the benefit by indulging our nostalgia - I'm full of admiration)
I distinctly remember buying a 1GB hard drive in 1992 for $1300 via an ad in Computer Shopper. And that was a really good price!
I buy Computer Shopper for the articles - said no one ever.
Bob and Alice was a great column. There were some programming articles too, I learned my first compression algorithm from one!
I used to buy the British edition at least partially for the articles, but it could be a damned hard slog finding them.
In fact a lot of the computer magazines tended in this direction - presumably the consolidation around the PC and the commoditization of it ate into margins as they all focused on the one (albeit growing) narrower topic.
By the by, I often joke that I buy The Economist for the pictures (not true, but the captions are often very sly jokes).
Including a regular Linux column by Charlie Stross before he started writing novels
Since it's been so long, I can't remember, but were there even articles - or was it just 999 pages of classified ads sold to the highest bidder?
I have vague memories of counting the number of content pages against the advertising pages once, and it was at least 80% advertising.
They had to have articles to qualify as a real magazine, probably for reduced mailing rates.
Most of the articles were product reviews, but I bought it for the BBS listings.
That must be the Holy Grail of publishing. You're buying it for the ads. So nobody complains that it's packed absolutely full of ads. And they're selling it to you. That's amazing. (And it was true! We did want to see the ads!)
I think I got my 1st custom PC from a vendor in CS. It was a Pentium 1 166Mhz MMX / 8MB RAM and 1.2GB HDD.
This is the box that I wrote a lot of VB6 code on for AOL, pirated my 1st MP3 (Deftones My Own Summer), and found my favorite SNES game (FF4). I didn't even know emulators were a thing!
With windows 98 and with 8MB of ram, i really learned about startup items.
There was a GE capital nearby that had a dumpster full of computer parts. I ended up finding some 8MB EDO sticks in the dumpster and it was the 1st time my computer would idle without trashing the HDD. 32MB was epic.
UK Computer Shopper was awesome in the early 90s. Fat wodge of newsprint covering everything from ZX Spectrums to NeXT stations.
I think I still have my copy of the issue where they reviewed the Sun IPX, ghods I wanted one of those...
> The cheap, cheap paper is a nightmare to run through a scanner
Not to mention having a faint mirror image of the other side. You need a dark background and a lot of light on the front to catch as little of it as possible.
And, then, one could add some smarts to try to align and subtract the image of the other side of the page to get an even cleaner image.
Now, just imagine if they kept an archive of their photo plates (or, better for later issues, PDF copies). By 2009 that's the very least one should have.
I loved Computer Shopper. I regularly bought the UK version in the 90s just to gawps at all those adverts and dream about buying my ultimate PC.
The regulars, including Don Lancaster, were great. I would drool over so many articles that I barely understood, and one of my goals for the next month were to get caught-up on some part of this tech that had escaped me.
Computer Shopper was fantastic for guiding me to complement my more formal studies.
OT: Does anyone else still use 6502/65c816's and TMS32C32's?
I was a Commodore guy in the 80's and 90's so didn't go for Computer Shopper as often as the rest of you, but did buy the occasional copy.
Having said that, today when I encounter the phrase "high capacity magazines" in the news, I chuckle because firearms isn't what first comes to mind.
I seriously thought about bidding on that ebay auction when it came up - I'm glad it went to a far better place! Good luck to him indeed - growing up in that time Computer Shopper was the logical progression from the Sears Wishbook.
Someone's gone and done it. My god.
You crazy bastard. I love you.
I wonder how feasible it would be to train an ai model to correct scanned pages, to avoid having to debind the magazines.
Uhh, to get the training set you would need a large collection of debound pages
The site wasn't prepared for the incoming visitor tsunami, though.
Glad someone is doing it. Passion projects drive tech.
Admirable effort. Thanks for preserving history.
What is the value of digitizing computer shopper?
Apart from nostalgia, it contains a staggering amount of information about historical PC pricing and hardware availability that (to my knowledge) isn't really documented anywhere else. Sure, Wikipedia can tell you the official launch date of the Intel Pentium, but Computer Shopper can tell you when the price actually dropped to something reasonable, when cheap motherboards for it started pouring out of no-name manufacturers, when vendors stopped bothering to advertise 486 hardware, etc..
Thanks for the genuine answer.
Read the room, and by the room, I mean this thread.