Computing related collection of stuff

Quoting the Microsoft Service Agreement (which seems to be obligatory for those using Windows Live applications):

4. How You May Not Use the Service.

In using the service, you may not:
use the service in a way that harms us or our affiliates, resellers, distributors, and/or vendors (collectively, the "Microsoft parties"), or any customer of a Microsoft party;

Unless there’s a more precise definition of "use in a way that harms", that just won’t work.
Is it "harm" if I merely work on some project (commercial or in my spare time, open source or not - doesn’t matter) that interferes with the business of such an affiliate etc.?

I guess Phoenix - one of the big guys in PC BIOS development - might be somewhere in that list - their software interacts in some way with Windows, and they probably coordinate their efforts.
Phoenix probably don’t like linuxbios too much - oops, I worked on that, and used e-mail while doing so. Would this clause prevent me from legally reading and sending linuxbios mails with Windows Live Mail?

My Ogg player broke when I tried to install its software on my XP machine - it seems that the software tried to update the firmware (without asking me) and failed for some reason.

A friend of mine is here for a visit and for fun, we plugged the stick in his vista notebook - and it instantly found some "STMP3500" "player recovery device". After searching for STMP3500, I’ve found an XP driver (different player vendor, but it worked perfectly) which allowed the firmware updater to do its magic: and now it works :-)

I just plugged my CBM1541 drive into the parallel port of my PC (with the appropriate adapter cable, of course), and it just worked. Even the very good CBM computer emulator suite VICE knows how to directly access it!

Now I only have to copy those 200 floppies I have at home and can dive into some nostalgia again :-)

I finally got a response to my inquiry about the Qt licensing exceptions for opensource licenses. I doesn’t look confidential, so here it is:

We will try to add the CDDL, but it may need to wait for a later
release. The next package is already going through testing, so I think
it may be too late to add it, but we will try.

 

I’m not sure how adding a license affects testing a release, but that’s their business. It still sounds great! Thank you, all the people at Trolltech for a wonderful c++ development library and supportive licensing behaviour!

I encountered some problems reading one audio CD into a flac file (I like to keep them in a one-file-per-cd format, with integrated cuesheet, etc) - not even good old trusty cdrdao on win32 managed to read it (but did just fine on DragonFly).

Searching a bit, I found that the ASPI driver (which provides the interface to lowlevel SCSI-and-similar devices for userspace applications) requires some flags in the registry. cdrdao even has the necessary changes in its cvs repository. Just apply, reboot (took me a while to remember that) and it works (for me).

Now, while foobar2000 is a nice player for my cd images, its ripping routine relies on the default TOC - which is the one in the latest session. so I’m back at scripting cdrdao and flac to create proper images (using the first session’s TOC, and fixing it up - which requires a local patch by me - like a red-book cd audio player would do)

So one of my next steps in that regard will be doing my own cdrdao build (with my own patches, etc), and creating a small rip-flac-freedb interface. I guess, that’s a good way to learn about win32 development, too.

After 8 Years of UNIX desktop (where not much changed - sure, glitzy-shiny features in the dozen, but still lacking in the usability and integration department), I bought a Windows license.

No Operating System without some weird corners, so I think I’ll collect such issues here (while looking for solutions). Sooo…

Episode 1: Try to delete a movie in a folder you just opened a few seconds ago. "This file is in use".. That thumbnailer/metadata extractor stuff again, with no obvious way to get rid of it (or to teach it how to read that file in a non-blocking way)

The "get rid of it" part is easy: regsvr32 /u shmedia.dll

Yay for GNU Freedom. Recently I ran over a discussion where there were some concerns about the viability of using "binary blobs" in an implementation of the AES algorithm.

Sure, there’s a table in it, 256 elements wide, with some weird numbers in it. What is that table? It’s an S-Box, one of the central pieces of every algorithm that’s based on the Feistel cipher. AES is such an algorithm.

When looking in the the spec (FIPS-197), you can see (on page 16) that they’re defined just like that: a table of values to use. Where do this numbers come from? Probably some NSA office, where some mad scientist (just like you’d image, probably) took the hints from page 26 of the proposal (or not) and shuffled a box with 256 numbers, and carefully placed them in a 16x16 square (think of lottery)…

Seriously, those guys tend to know things about crypto that they prefer to not talk about. And they knew 1975, 15 years before the rest of the world figured it out, that random placement of those numbers in that square is not a good idea, as proved by their work on DES.

So please, those numbers are necessary, and they’re necessarily in this order (as otherwise you’d get a different algorithm), and they very likely in exactly that layout for a good reason. Reading up on crypto algorithms even tells you what to do with them, and the current form and shape of such arrays is more than enough for modification…

But why would you want to modify it?

I continued work on hg2mtn (which managed to import the whole onnv-gate repository just fine, stats will come later), and except for one issue, I’m feature complete: while I track renames now, I have to synthesize directory renames. This will work by looking for each rename if the source directory is getting removed and the target directory newly created. If so, assume that a rename from source dir to target dir takes place.

This might give some wrong decisions, eg. in this scenario: a/a is moved to c/a, while the other 100 files (that are sorted behing a/a) in a/ are moved to b/.  The proper rename would be from a/ to b/, but a/ to c/ is chosen. I could do majority "votes" and things like that, but they’d quickly get too complicated while still staying a heuristic.

The only issue that happens in that case is that a new file in a/ from some branch gets moved to c/ instead of b/ (where it likely belongs to) - but that’s easy to solve and still better than the current scenario where you’d have to move the file out of the deleted directory before merging to avoid a conflict.

After I implemented that one, I only have to clean up some more (I started already) to make a stable release, yay!

Darren Reed’s thoughts on the opensolaris community buildup - very interesting read!

It seems, that the "architects" came to similar conclusions already, given that the community/project stuff was reorganized a while ago. Unfortunately it’s still very rigid.

Of course, it doesn’t matter to Project Independence - we won’t ask for a slot on opensolaris.org, so we won’t have to mess around with their idea of "community".

Today I worked a lot on hg2mtn, a tool to import hg (mercurial) repositories into monotone.

The result of it is a first release that’s barely useful, but contains all the logic necessary to do it - what’s missing is userinterface, stability and general niceness.