The early days of CD-ROM
A colleague came in clutching a copy of the Microsoft Dogs CD-ROM. Another colleague immediately asked, "What's on it? Windows 1.0? Word 2? Bob?".
One anniversary unlikely to be marked within the bowels of Redmond is that of Microsoft Bob. However, a retired engineer has taken to YouTube to confirm that the unloved interface did indeed enjoy a second life as digital ballast for Windows XP. Having spilled the beans on Task Manager back in May, the former Microsoft …
Other than being used to determine if oem or retail versions by the installer.
And IIRC some early CD roms (cd-r's were not ubiquitous then, many a Friday night I funded through upgrading family multi media centres to facilitate disk burning) had a fit if the CD wasn't full so probably needed just in case someone cobbled min specs together to get the windows eXPerience (bsods, reboot loops, telly tubby desktop, plug and pray, hours of fun fighting for your pci modem to be recognised only to fall over come cumulative update time, deafened and startled by the startup and shutdown noises, so user friendly skiddies could pwn the world with a few lines of vbscript... )
Plus I believe at the time full payloads were more economical to press as cds, which was why the iso vs installer images still differ in size...
"Why?"
Besides increasing download times for ~56K?
1. If the dummy data was at the start, the read speeds would be increased.
2. Similar to 1, back then jitter was still a thing and it was always bad at spin up.
3. Again similar to 1, back then positive offsets could be very high, missing many samples.
4. Some pressers required a full image to construct a master (I never found out why).
FWIW, the readable side (the grooves of data), are pressed into the plastic. The film is only for the laser and can be replaced (I'm still curious if a method can be devised to read a CD from the top down, thus skipping resurfacing).
Completely off beat, dummy data was essentially how Sega protected the Dreamcast with GD-ROM (dummy CDDA tracks... essentially ;-/).
Priti Patel is now the purveyor of that approach. See Dave Brown's cartoon that appeared in the Independent recently.
No, but as Chen points out in his blog post, even as far back as XP it had CryptGenRandom, so it's trivial to create a program that will dump an arbitrary amount of crypto-strong pseudorandom data. In fact, I think you could have scripted it with CAPICOM and WSH, using either Visual BASIC or JScript. (CAPICOM isn't in modern versions of Windows and I don't have XP handy to check.)
Based on Plummer's comments quoted in the article, he 1) wanted to use the Bob blob for nostalgia, and 2) did a bunch of cargo-cult crypto that had no useful effect or meaning. Any decent symmetric encryption algorithm should serve as a PRF in the first approximation, so all that stuff about "several passes of different encryption tools and techniques" is nonsense.
On the day a certain application (as we called them then) was due to be released, the programmers found they could fit the installation one one floppy rather than two as planned. Only the sticky labels had already been ordered, so they padded out the installation with some random data in order to spread it onto two floppies.
Wow!!! A good ending.
At one company I worked at I would have been penalized. I remember scheduling 2 weeks to find a bug no one could track down, found a work around in half a day, got penalized with a bad review for mid-estimating the time, quit and worked for the really large company in Cupertino.
> I remember scheduling 2 weeks to find a bug no one could track down ... [and] got penalized with a bad review for mid-estimating the time
It has to be a special kind of twat-ish company to (a) know that they need to fix a bug but (b) will set a time limit on doing so. What were they going to do? Withdraw the product if it couldn't be found? Or let each of their developers have a go, one after the other, two weeks each?
I worked for a certain very large British Bank about 15 years ago who penalised their project managers if their planning was out. If we finished work earlier, we also padded it out until the planning was met. If we didn't have enough time, we finished anyway. The planning was the most important part not the actual work.
An approach that, to anyone with an ounce of common sense, would ensure that all project estimates given would always be a worst-case scenario and would almost always be proven entirely correct. Thus ensuring that the maximum amount of time was wasted.
I've worked for companies like that :)
"The planning was the most important part not the actual work."
This applies to procurement processes at British universities too - the process is far more important than actually achieving value for money (and the suppliers are fully aware of this - some vendors INCREASE prices over posted retail when you login to place orders)
Brings back memories, but in the 1970s, and with ROMs rather than floppies.
We typically shipped early builds of a product on PROMs (programmable ROMs), while waiting for the cheaper masked ROMs to come in. The program image took 6KB, so 3 2Kx8 PROMs (or six 2Kx4 PROMS for the more expensive but easier to get bipolar ones).
All went well until the Masked ROMS (two 4kx8) arrived, and we got an urgent "Line Down!" call from the factory, saying _ALL_ of the new ROMs were defective. Turns out that rather than verify the checksums, they had checked and found that the first few bytes of the first ROM were _BLANK_!!! So the sky was falling.
I managed to talk them off the ledge, but to forestall a repeat performance, I made a revised image that filled the first 2KB with a character set I liked (Hey, it wasn't blank...) and later used one of those chips in a homebrew terminal.
I managed to talk them off the ledge, but to forestall a repeat performance, I made a revised image that filled the first 2KB with a character set I liked (Hey, it wasn't blank...) and later used one of those chips in a homebrew terminal.
That brings back memories.
I had Kaypro "portable" CP/M computer which had nordic character set. What they had done, was replace {}[]\| with the nordic characters. Made programming a bit painful. So I read the contents of the EPROM. Edited the character map to "re-create" the braces etc. I then burned it to a new EPROM. Finally I soldered the new character map EPROM on top of the old character map EPROM (well, a copy of the original as I wanted to retain the original in case anything went wrong) with the exception of chip select pin. I then installed a switch to switch CS between the two EPROMs and could flip between character sets in instantly.
@down not across
A nice easy fix :-)
Thanks for the memory .. You reminded me of a RAM upgrade I did in a similar manner by doing the same approach but adding a 74xx series gate on top in dead-bug mode to add the right decode lines to form the separate CS pins.
Aren't finance departments wonderful?
Option 1:
Programmers: We were planning to have a unit distribution cost of two floppies, two labels, an envelope, and the initial documentation. We met that target.
Finance: Well done.
Option 2:
Programmers: We were planning to have a unit distribution cost of two floppies, two labels, an envelope, and the initial documentation. We managed to reduce that to one floppy, but still two labels because that was last minute.
Finance: You're wasting money on the extra labels! Even though we were already planning to spend that and you saved us money on the floppies! Let's fire somebody.
"Plummer also worked on the first incarnation of Product Activation for Windows: "A necessary evil. Had to be done. Sorry about that."
Was a real shame. However it did get me to move my butt and migrate to FreeBSD. Possibly not as convenient or user-friendly as Windows XP but then again, I very much prefer it to Windows 10. Open-source desktop environments certainly didn't get better. Windows just got worse at a faster rate.
Since the WPA has been completely bypassed these days with KMS emulators like KMSpico, perhaps Microsoft would consider removing the crippleware again? Since it serves very little purpose other than alienating legitimate users who run offline labs or refuse to condone DRM?
DRM is only for casual piracy anyway. Anybody devoted to finding a way around it, will. Most likely those devoted enough to defeat the DRM to that wouldn't have paid anyway.
Even so, the current methods only work because they could do better, but its still a balance for how much they want to piss off their legit users and how many hoops they have to go through vs. preventing average small business user from buying one copy and putting it on all their computers.
If they let the flood gates open for casual pirates, they'd see a huge drop in revenue.
I'm not so sure about this. Many small independent businesses do tend to use cracks for their internal software (in particular Photoshop and Sage clients). These businesses are usually compliant with licenses but did not want to open up the ports required for activation or the DRM encumbered software does not support a proxy (this is the most common!).
So these companies "pirate" and still buy licenses in case an auditor comes their way. As a contractor I have experienced this in the majority of companies in the UK. Usually everyone is fairly hush hush about it.
My baby was talking in her sleep
She said, "Bob?"
I said, "What?"
She said, "Bob?"
I said, "I'm Bob!"
She said, "No, I mean a different Bob."
And so I asked her the next day
I said, "Jane?"
She said "What?"
I said, "Who's Bob?"
She said, "You're Bob!"
I said, "No, I mean a different Bob."
My baby was run over by a truck
It was a Dodge
As she lay dying, she called for Bob,
I touched her hand and said "I'm here!"
She said, "No, I mean a different Bob."
Having recently seen Microsoft Bob in action over at the LGR YouTube channel, I found that I had to severely adjustment my opinion of this software. Most of all that it isn't terrible. The fact that there were at least half a dozen similar solutions out there around the time showed that the concept made sense, and for a person new to PCs even today, the use of a UI that matches what they already know isn't a crazy idea.
As someone who has had to guide many a family member and family of friends (etc. etc.) through software installations, basic computer usage and driver installations (insert hatred for WinModems here), I'd rather give them something along the lines of MSFT Bob rather than spend another untold hours explaining to people what the heck a 'taskbar', 'start button' and 'mouse cursor' are, or what these 'programs' are and what this 'window' is in that they open.
Yeah, I like Bob well enough :)