Slashback: Bundestux, Kerberos, Blizzard

Slashback tonight with several updates and amplifications, starting with a nice report on the current state of the effort to put Linux into the heart of the German government, but also bits on Starcraft, cleaning up UNIX config, and Kerberos.

This deserves a hearty 'Jawohl!' DocSnyder writes: "Since the Bundestux campaign started collecting votes in favor of putting Free Software into the German parliament (Bundestag), more than 25000 people have done so. A lot of online discussions - in addition to Heise News and Linux-Community.de, even some Bundestag parties have put up their online forums - are very active to share user experience about GNU/Linux and Free Software. (Sorry for most of the linked sites speaking German, it's simply too much to translate at once.)

After several open letters and press releases have been exchanged between lobbyists and politicians, some information about a research performed by the German company Infora appeared on Heise News (english version), recommending an all-Microsoft infrastructure with the exception of some security-critical services like e-mail. The detailed paper is still not available.

An internal test (english version) between the Bundestag administration, SuSE, IBM and Microsoft confirmed that GNU/Linux and Free Software are in fact ready for the Bundestag's IT infrastructure, yet the testers don't like the copy&paste method used by KDE and recommend Windows for the desktops.

Last week, the Bundestag members (MdB) Jörg Tauss and Hans-Joachim Otto have been invited by Heise for an online chat with the community. While Jörg Tauss is a clear supporter of open standards and Free Software, Hans-Joachim Otto takes the internal test as well as Infora's research as primarily relevant for the coming decision.

On Saturday, MdB Uwe Küster summarized some details in an interview. He considered the decision - officially due Feb 28 - as almost finalized. The solution would show GNU/Linux on most servers, Windows XP and Office XP on the desktops, keeping proprietary data formats and lock-in interfaces up to the next upgrade cycle, which in fact would have been problem number one to solve.

All in all, the community has provided lots of experience, ideas and solution paths which finally seem to be largely ignored in the decision finding process towards the successor of a homogenous Microsoft Windows NT4 infrastructure, which has to be replaced until 2003 when Microsoft will no longer provide support for NT4."

That's a lot of cleaning up to do! maffew writes "A lot of feedback and ideas have been flying around since my article How to fix the Unix configuration nighmare was featured on freshmeat and slashdot. So we've created an ongoing web site and mailing list for people to continue discussing, organising, and hopefully in the end coding. It's all at unixconfig.sourceforge.net.

Meanwhile here's a link to the permanent home for the nightmare article. This is where I'm making revisions and adding links."

Raise your hand if this would mean seeing it for the 4th time ... Chris Brewer writes "In case you've been living on a different planet, The Fellowship of the Ring picked up Five Baftas, the British equivalent of the Oscars, including Best Director, Best Film, and Peoples Choice. During a live interview (Real only) after the awards, Peter Jackson announces that a preview for The Two Towers will be shown from the March 22 screenings of The Fellowship."

At long last ... something? If you've followed the strange relationship Microsoft has had with Kerberos, you may feel grateful to the anonymous coward who writes: "It would seem that Microsoft is granting the world a royalty-free, non-exclusive license to implement their Kerberos extension."

Here's some comfort for Starcraft players. An Anonymous Coward writes "As stated on Blizzard's battle.net service, the latest Starcraft patch supports UDP play, so some of the compelling reasons to use bnetd have been addressed. Whatever you may think of Blizzard and the DMCA, at least it shows Blizzard is listening to its fans."

Comments

MS Kerberos, a corporate culture of wrongness

[WasterDave]

Yipee! They published their wire protocol:

"All data is encoded as little-endian."

Oh, god. Look, since the start of time itself binary data on the net has been big endian. No, you do not know better.

Head->table: Bang! Bang! Bang!

Dave

Re:MS Kerberos, a corporate culture of wrongness

[AaronStJ]

Oh, god. Look, since the start of time itself binary data on the net has been big endian. No, you do not know better.

Um, I don't mean to flame, but why does it matter? Byte ordering strikes me as rather arbitrary. Except for the fact that you probably want to keep any new standards consistent with the existing dominant processors, which seem to be little endian. At least, I for one am annoyed at having to call ntohl() every time I want an int I pulled off the network to be usuable.

Re:MS Kerberos, a corporate culture of wrongness

[taniwha]

Byte ordering strikes me as rather arbitrary

actually why it's an issue at all is shrouded in history ..... Arabic numerals were originally written in - well - Arabic which is a right-to-left writing system ... however in the original numbers were written LSdigits first, which makes sense since a culture of traders back in those days mostly did addition, which is done LSdigit to MSdigit .... it seems that some monk in Spain (where the Arabic numbering system was adopted when the Moors were driven south from Spain) ... didn't really realize what he was doing an copied the Arabic system into Roman scripted languages (ie left to right) without reversing the order of the digits - the result is the mess we're in today - a scripting system where we think of text being left-to-right, but numbers being right-to-left (except we're so used to them we don't realise this). This means that there's no real culturally 'natural' way to handle byte ordering for us (if our first stored program computers had been invented by native Arabic speakers they'd all be little endian and it wouldn't be an issue)

Microsoft and Kerberos

[hillct]

I remember vividly my discussions with microsoft personel at the Win2K launch even in North Carolina. We were debating the validity of adding data to optional fields in a Kerb ticket which would effectively prevent a ticket issued in a unix realm from beung useful in a Windos Kerb realm, but not the reverse.

After filtering out the marketoids who repeatedly insisted everything was fine, a couple engineers conceded that the implementation was broken. It;s interesting to see Microsoft try and sell this as an extension that others shoupls implement and use. Unfortunately, this is yet another example of the effect of monopoly power.

'We support the standard but if you want to access our systems you need to implement the standard our way'

What a sham.

--CTH

Copy and paste of all things...

[Andrew Coles]

...yet the testers don't like the copy&paste method used by KDE and recommend Windows for the desktops.

Choosing a desktop on the basis of a copy and paste model. I thought people got their priorities wrong but this takes the biscuit. Copy and paste vs free and more stable...

Just out of interest - how easy would it be to port a windows-style copy and paste model to KDE? I thought the KDE UI was relatively customisable in this sort of area so implementing such a feature would be relatively easy. Then again, I could be completely wrong.

Re:Bundestag

[jezreel]

Oh yeah... just read it.

The reason seems to be money.... Should've thought about this earlier

Microsoft bows to outside pressure?

[AndyDeck]

I'm practically speechless here... Microsoft has actually relinquished a proprietary lock.

I am neither expert enough at Kerberos nor Samba to know if the above-referenced web page (Here in case you missed it) is truly sufficient for interoperability, but it sure looks like it.

And the critical language is at the bottom:

Microsoft grants you a perpetual, nonexclusive, royalty-free, world-wide right and license under any Microsoft copyrights in this specification to copy, publish and distribute this specification, and to implement this specification in your products.

and

Microsoft is not currently aware of the existence of patent(s) and/or pending application(s) that are essential to the implementation of this specification. However, if Microsoft becomes aware or has any patent(s) and/or pending applications that are essential to implement this specification, Microsoft will grant you a royalty-free license under applicable Microsoft intellectual property rights essential to implement this specification for the sole purpose of implementing this specification. Microsoft expressly reserves all other rights it may have in the material and subject matter of this specification. The licensing commitments made hereunder do not include any license for implementation of other published specifications developed elsewhere but referred to in this specification.

Translation: You can use this spec in your products. It's not covered by any of our current or pending patents, and even if it is, you can still use it royalty-free.

Other related specs are not rendered licensed or royalty-free, so they MAY have kept a loophole - but this looks sincere so far.

Amazing news, really.

Re:Microsoft bows to outside pressure?

[Soko]

I think you can thank one Vinod Valloppillil for exposing the larcenous behaviour of the Microsoft Marketing Dept. - and thier influence on Microsofts coding practices - for this one. Much news about Microsoft has transpired since then, and most of it very, very bad for them.

/imagine: a half bald, bawling marketdriod with a pile of his own bloddy hair at his feet, more in his fists at the side of his head, shouting "NO! We could of owned it all! NOOOOOOO!". Now smile. :-D/

IMHO, it was the leaking of the Halloween Documents that had the most devestating effects on Microsoft. Those very same documents also give reason to still be guarded with our support.

The timing of this is also suspicious with .Net on the way. From what I've read, most big comanes are telling Microsoft that .Net is OK, but thier keeping all thier data. Ergo, what if Microsoft wants to use Kerberos authentication from your server, whatever it is, to allow .Net services in? It may be needed for thier business model to actually work. Hmmmm.

To paraphrase a famous Trojan :
"I fear Microsoft even when they come bearing gifts".
I'm pleased, but will still be very wary all the same.

Soko

Read it again, because they *have* to.

[Bob_Robertson]

Since WinNT is closed source, the only place to get service and support is from Microsoft.

Since Microsoft will not be supporting NT4 past 2003, they are wisely getting their house in order before they run out of time.

Everything has to change anyway, so they are entertaining the idea of building something better. Unfortunately, the argument that they don't have to replace the hardware at all wasn't good enough.

So rather than spend the money on consultants to use existing hardware, and free software, to rebuild their network, they are going to spend the money on consultants, new hardware and new software to retain the same functionality they have now.

And in a few years, when Microsoft stops supporting XP, they're right back where they are now.

That's why it's called an "Upgrade Cycle".

Bob-

Blizzard

[Seclusion]

"Whatever you may think of Blizzard and the DMCA, at least it shows Blizzard is listening to its fans." And it only took them 4 years to get around to this issue. Not that I'm complaining much, blizzard is better then some other software company's when it comes to patching games after sales for them have dropped. Still, I have to wonder if it's worth supporting a company that represses the people who actually buy their software in the name of piracy protection.

Starcraft

[Anonymous Coward]

Actually the UDP support has nothing to do with bnetd. Starcraft was ported to Mac OS X, which doesn't have a working IPX protocol that they could use. So rather than try to graft IPX onto the Carbon version of Starcraft they created a new UDP version for LAN play with OS X, then added UDP to the other versions afterwards.

There were a few posts to insidemacgames.com's forums by the Blizzard techs who made the patch.

Listening to which fans?

[Kasreyn]

"Whatever you may think of Blizzard and the DMCA, at least it shows Blizzard is listening to its fans."

Oh? Which fans might that be?

I think I can speak on behalf of D1 players everywhere: over 5 years on the clock and still running. Where's the patch for the dupe bug, Blizz? Oh, what's that you say? There's NOT a patch for the most egregious bug in the game YET? After 5 YEARS? And don't even get me started on all the other bugs that would be easily fixed if they gave half a rat's ass.

Hm. So much for the fans.

-Kasreyn

Re:Listening to which fans?

[Bowfinger]

Be reasonable. D1 was great in its day. I lost thousands of hours to it. But D1 is a dead product, superceded by D2. If you liked the original, you will probably like D2 even better. It's the same basic game, but with far, far greater variety and replay value. Of course, if you didn't like D1, you may not like D2 either. To each his own.

There is no replacement for StarCraft. There doesn't appear to be one on the horizon. I think it is remarkable that a company like Blizzard continues to support and enhance it. How many other companies are still provding free patches for five-year-old games? How many other companies even provide bug fixes for old products like this with little sales potential?

I too am concerned about Blizzard's actions with respect to bnetd. I understand their legitimate need (and their right) to control the spread of the Warcraft III beta, but they overreacted. I hope this is just an aberration. Too many companies seem to have run out of fresh ideas of their own, so they use the legal system to suppress fair competition. It would be a shame if Blizzard has joined that list.

Yes, Blizzard has problems. If you look at their overall record, I think Blizzard is still one of the good guys. There don't seem to be many of them left. Give Blizzard a little slack, at least for a while longer.

But that's just my $0.02 worth.

UnixConfig (tm)

[KidSock]

I was ready to post my message to the UnixConfig message board but apparently I wasn't logged into sourceforge. I think I'll just post my comments in response to the documents prepared there here instead:

--8<--

A core system would handle parsing, verification and storage of text-based configuration files in one or two basic formats.

We cannot do this. We must be able to handle arbirary file formats. There is no way we will get anyone to change the format of Samba's smb.conf, Apaches truly arcane httpd.conf, or DNS zone files for example. We *could* standardize on a uniform in memory representation but I'm not in favor of that either. I think we have to go all the way up to the API level (e.g. int exports_add(const char path, int flags, ...,{SMB|HTTP|NFS|...}).

The master copy of the configuration is always left in the native text files (in /etc and ~/.*). This is where linuxconf falls down, it starts keeping its own copy of the configuration, which means if linuxconf takes over your system and then later something stuffs up, it's difficult to edit a text file manually without losing linuxconf completely.

Absolutely. The confuration files *are* the database. On a separate front, we might provide an idealized open-ended application configuration library for assisting new development but I think there would have to be some weight behind the main front before developers would even consider it. That might also give us the opportunity to normalize on a few file formats (e.g. scanf, WINI, XML).

Another option is to allow plugins to handle how the data is stored.

That's a goodish idea but there are interfacing issues. By "plugins" are you suggesting one could write their parser in C or C++ or Perl? At what point do you normalize on a common language? Keep in mind this has nothing to do with *file* formats.

In order for some features to work, it might be necessary for application developers to switch to the use of the configuration manager for their internal routines.

We cannot do this. We must transparently manage data within the configuration files of the applications themselves. There is no way in heck we'll get app developers to convert. Their intrests are far more important in their mind (and they're probably rigth).

A key element would be the configuration format description file. This would list the configuration options for a given piece of software, giving for each one the name, type (boolean, list, string, filename, internet address, etc.), options, category (for sub-sections within the config), and help text (short and long).

You'll end up with a glorafied property editor and that's not what you want. What I mean by this is that you do NOT just want to map configuration options within application config files to the configuration options of whatever tool we're talking about. This is one of the greatest failures of UNIX confuration tools. It would be far more effective to isolate and the concepts associated with changing the behavior of a system (or group of systems) rather than just mapping check boxes to booleans and selects to lists. The KDE runlevel editor is a spectacular example of this failure; it does not isolate the concept of what it means to change the initialization behavor of your system.

For example, rather than writing configuration screens for Samba, Apache, Pro-FTPd, and NFS exports, write an "Exports" module that handles all of them uniformly. They all do essentially the same thing; make a portion of your filesystem available as a network service. Similarly, instead of having a PPP dialer, make a module that controls your "Network Interfaces" (RH has largely done this working PPP into network-scripts). Again, isolate concepts rather than parameterize configuration options.

Re:UDP is an Internet Protocol, right?

[snilloc]

What, you mean exactly like Kahn did for IPX before it mysteriously vanished from the internet? Kahn

If anybody can actually tell me what the hell happened to Kahn, you get a cookie or something. At least one of the Stargate Network guys must read /. !!

Re:Why do all the games seem to use IPX for lan pl

[Anguirel]

Because, for LAN games, IPX is a better protocol. It bypasses a lot of routing issues that happen otherwise, and is simply faster. Try running a game that supports both (loke Age of Empires 2) via TCP/IP, and then one via IPX to the same person. You'll experience a lot more lag on the TCP/IP game, in general. I'm not certain of the reasons behind this, but I know that IPX has a 'guaranteed' bandwidth that TCP/IP lacks since it can only be used in LAN settings.

And... I though Mac did support IPX...

Kerberos - not all released

[dirkx]

So we hear:

"Microsoft is granting the world a royalty-free, non-exclusive license to implement their Kerberos extension."

Hurray! But... It is jus the license and doc's for half of their extensions: the part which does group enumeration. Which was already understood anyway.

The real beef - i.e. the domain controller specifics - are still as closed as ever. And according to the presentation at the RSA conference last week - are going to remain so.

Congrat's to slashdot for picking it up just as the spinmeisters intended :-)

Dw.