Bobulate
Home [ade] cookies
KDE Software Compilation 4.4.0 and OpenSolaris
February 9th, 2010
Although KDE SC 4.4.0 isn’t officially released yet (or there’s no announcement on the dot yet), I’ve bumped the specfiles in the KDE4-OpenSolaris repository to 4.4.0 and kicked off a rebuild. That means that we should have new packages at about the same time that the major GNU/Linux distributions have them. The release candidate was pretty stable on OpenSolaris as long as you didn’t (re)build any parts of it while logged in, but we’ll probably be hearing enough bug reports in the next few months for OSOL. You can report problems to KDE’s bugzilla with the OS set to Solaris, or use the OpenSolaris bug tracker and choose one of the KDE components there.
Of course, the whole future of OSOL is a little .. fuzzy. The Oracle / Sun takeover has caused a fair amount of shake-up in many of Sun’s projects, but OSOL has not, as far as I know, been either committed-to or deprecated. We’ll see.
In the wake (er, .. the bow shock wave, since I’m getting ahead of myself) of the release, Jos asks for something offensive and the KDE website has been re-vamped. Especially the latter is impressive, for breathing new life into the somewhat moribund site. I believe the call for help still applies as the rest of the content is polished and brought up-to-date.
Bits of Privacy
February 8th, 2010
The organization “Bits of Freedom is a Dutch NGO Privacy watchdog. It has been re-vivified in the past years. It hands out semi-regular Big Brother Awards (say, isn’t the estate of George Orwell going to wake up at some point?) for the worst offenses against privacy in the Netherlands. Worth reading if your Dutch comprehension is ok. BoF has garnered some mainstream media attention as well, which offers hope that privacy isn’t completely a lost cause in this country. [Current schemes of nationwide chipcard travel and road-pricing deny that hope] There’s also a “Winston Prize” for someone fighting for privacy, which went to the Euro-MP Sophie in’t Veld. It took me a minute to figure out the “Winston” name, but since I’d read 1984 (it says so right there in my book list) the light dawned eventually.
Anyway, cheers to BoF, and keep watching over your own privacy.
O noes, book list meme
February 8th, 2010
I was reading some hip and literary Swedish blogs (no, actually I found this stuff on some tech blog, but swimming down the attribution links brought me these far more acceptable references) which talk about a list of books one might be expected to have read. The list has apparently been circulating for some time, witness the lonely librarian, and continues to be passed around, for instance on the symbian diaries (I can’t find any Symbian content there, though). It might be that the list stems from this story (from 2007?), but I’ve seen the Guardian reported as the original source as well.
The list itself (in the form in which it’s being passed around right now) is a little odd, since there’s some strange entries: the Bible (which version? does it matter? with or without the apocryphal books?), the complete works of Shakespeare (zounds, that’s a lot, but it also duplicates Hamlet much later in the list). The list mixes classics with much more recent work and includes some real cruft (IMO) as well. Anyway, I thought that publishing the list would be far too space-consuming and not interesting on a technical blog, so I wrote a python program to produce HTML and plain text renderings of the list, as well as a compact text representation using the initials of authors and titles and a base-3 representation.
You can run the program with something like python booklist.py text to get plain text output; alternate modes are html, code which outputs a kind of geek code, and short which produces a decimal integer. The program comes with an unread database, so it will print an all-unread text representation, or the integer 0. The integer corresponding to my own reading of the list is 472899411102988434671899921134218056756239761136, and you can add that as a second parameter to get the output showing what I’ve read; in code form that’s:
*PaP-JA *TLotR-JT *HPs-JR *TB-: *WH-EB *NEF-GO *GE-CD *C22-JH
*CWoS-S *TH-JT *CitR-JS *M-GE *GWTW-MM *TGG-FSF -WaP-LT *THHGttG-DA
-CaP-FD *GoW-JS *AiW-LC *TWitW-KG *CoN-CL *E-JA *P-JA *TL,TWaTW-CL
*WtP-AM *AF-GO -OHYoS-GGM *AoGG-LM *THT-MA *LotF-WG -LoP-YM *D-FH
*SaS-JA -ASB-VS. *ATOTC-CD *BNW-AH *OMaM-JS *TSH-DT *MD-HM *OT-CD
*D-BS *TSG-FHB -U-JJ *ACC-CD -MB-GF *CW-EW *AoSH-SACD *TLP-ADS-E
-TWF-IB *WD-RA *TTM-AD *H-WS *CatCF-RD
To generate your own book statuses, you’re going to have to edit the program and insert a status (READ or WANT) into some of the tuples that form the book database. Then run the program with only one argument for the desired output format. Once you have the code or short output, you can post it and keep the booklist meme alive (albeit in a more nerdy form).
The program itself can be downloaded as source from here, and is under the GPLv2 or later (bear that in mind if you send patches for an interactive of Qt version, and please don’t berate my python style).
There’s clearly features missing from the program: there is no Qt interface (with PyQt bindings) and no interactive mode, which would make it much easier to generate your own booklist output. Abstracting the purpose of the list would also be nice, and I can see a Meme-Plasmoid (a memeoid?) somewhere in the future where you get the checklist of the day and can enter your results for that checklist, to share across the social desktop. O noes!
Scratch that
February 4th, 2010
I’m still sick. I still can’t hear properly, nor breathe, so as FOSDEM draws nearer (where I would man the FSFE booth, attend a few talks, and promulgate a business-friendly selection of Free Software licensing topics) I’m forced to re-evaluate whether I can go there at all: no, I am not going to FOSDEM. I’ll miss Paul’s bits on PIMping the KDE desktop, which is a particular shame. I’ll miss Michael Meeks, with whom I was looking forward to chatting about copyright assignment. I’ll miss Vincent Untz and the Dutch GUADEC team. I’ll miss Claudia and the KDE gang. I’ll miss the OSOL guys. I’ll miss the cute lovable blue Postgres elephant. And the weird devices at the OpenBSD stand. I could go on.
Instead, I’m going to stay home, eat lots of oranges, sleep a lot and hope for a better future. One where I can actually hold a conversation without phlegming anyone to death.
On a totally unrelated note, I’m reading “de telduivel”, a book on mathematics for kids by H.M. Enzensberger (English title “the number devil”, but the book is originally in German). It’s a fun book, and this time I’m actually reading it to the kids. Their minds do wander when we get to topics like fib(n)^2+fib(n+1)^2=fib(2n+1), but it’s inspiring for me as well. It’s been a long time since I sat down to try to do a little proof on paper, and since I don’t have to talk or breathe very hard to do math, it’s fun!
KDE SC 4.4 beta 2 on OpenSolaris
February 1st, 2010
The second beta release of the main collection of software produced by the KDE community, aka KDE SC 4.4 beta 2, was released last week. There are OpenSolaris packages available thanks to the compilation efforts of the Pavels; since we’re still not very good at publishing the packages in a standard repository, nor in managing smooth upgrades, you can find the packages here.
This set of packages includes Qt 4.6.1, which is a marked step up in stability; also Virtuoso support has been added if you build from source, but not in the packages. We’re still wrestling a bit with the C++ stack, as some things still use the older Cstd STL and “our” stack uses the newer Apache stdcxx4. In OpenSolaris, things are supposed to move to the Apache stack, but that takes some time and needs careful coordination, for instance because enchant plugins need to be moved over, but that implies that all of GNOME (the parts of it that use a C++ API anyway) also needs to be recompiled.
The first time I started up 4.4 beta 2 was a real eye-opener, from the new artwork to various bits of polish applied to the desktop (well, on OpenSolaris that includes things like “you can now start an application from the K-Menu again”, which was broken in all the recent versions). Since then I’ve started jotting down comments about aspects of the default installation that bother me — mostly little things, and it’s another indication that the packaging is coming together pretty well on OSOL that little bits that lack in polish are able to attract attention again.
Once I’ve got a more substantial list I suppose I’ll blog about them. That’s a little tricky to do, since many of them might be a “designed that way, but incompatible with [ade]” thing that I just don’t know about and I don’t want to come across as whiny or demanding (I’ll save that for licensing issues). And I don’t think that things like “Akonadi startup screen gets in the way” are worth filing a bug over, not until there’s some consensus that it is a bug. See also the “logout sound” thread on KDE core-devel right now, that feels similar.
Anyway, it’s good to see that the next release will be pretty solid, also on OSOL; if you want to take a peek, bother me at FOSDEM this weekend (just before the 4.4.0 release!).
FOSDEM schedule (my own)
February 1st, 2010
Like many folks, I’ll be going to FOSDEM this weekend, in Brussels. Thomas Koch posts his whole weekend schedule — gosh, that’s ambitious. I know I never get around to attending one tenth of the talks I might intend to see, simply because there’s too many people to talk to, things to demonstrate, chats to have and random other interesting things to do. Given the state of my health, I think I’m going to remove “excessive amounts of beer” from my list-of-interesting-things, but I’m still looking forward to Brussels Cheese during the social hours. Isabel Drost is giving a talk on Hadoop — that’s one I might actually make a point of attending, because otherwise I’m never going to understand any of her blog posts. I guess like I’m tedious about licensing, she’s tedious about map-reduce. We’ll see.
For most of the weekend you will find me at the FSFE stand near the front entrance. I hope it’s less freezing cold than down the other end of the hall where the KDE stand is, traditionally. I’ll be available for legal and licensing talk, mostly. Governance? Patents? All ears. For technical stuff I will need to escape from the watchful eye of my handler, Hugo Roy, and covertly discuss bits and bytes.
*sniffle* (2)
January 28th, 2010
Another week gone by an no end to my cold in sight. This is getting to be really annoying, as it both fragments my work day and makes me unreliable in responding to just about anything — so various jobs (both real and hobby) have been falling behind.
The hobby bits are easier to write about, so here goes: I wanted to produce a screenie of Krita on OpenSolaris, but it turns out that it hand not been built in my first packaging (neither had KSpread, for that matter). Turns out they require Eigen. Eigen is a C++ template-based linear algebra library. So I installed the Eigen package (also produced by the KDE4-OpenSolaris project: make no mistake, this is a small group with a big software stack to take care of). That’s where the fun starts — I don’t think anything else makes serious use of Eigen. Maybe some stuff in KDEgames or KDeedu, but obviously because I hadn’t had Eigen installed previously, they (if they are there) have been silently ignored.
So, Eigen. Since the KDE4-OpenSolaris project aims to use the Sun Studio compiler (so that the resulting KDE4 could be included in the base system and so that i t can re-use any other C++ infrastructure in the system), we need to make sure that Eigen works with the compiler. In the case of a giant and complicated template library, that can be a big effort. And I’m immediately stymied by the Matrix class, where the compiler complains about multiple definitions of every type in the Eigen generic interface (Base, Scalar, and another half-dozen). I’ve not even managed to strip the examples down to an example that demonstrates the problem, which makes debugging — or asking the compiler gurus for help — difficult.
Suffice to say that Krita screenshots are not to be expected anytime soon.
On the upside, the release of KDE SC 4.4 RC2 means that the other bits are nearing stability, which means we can re-focus on stability and performance on OpenSolaris for a bit. Since Pavel has set up various continuous builds on our crunchy build box (courtesy of Sun Netherlands) so tracking in future should work a little better — I hope we have fewer separate patches to maintain in future as more stuff gets upstreamed.
FSFE awarded medal for good governance
January 26th, 2010
The Theodor Heuss Foundation (German site, but the Wikipedia article is probably more informative), named after Germany’s first post-war president, is a foundation established to remember the political achievements of Theodor Heuss as an example for social commitment, moral courage and the dedication to fostering democracy. The foundation seeks “to bring attention to something, which has to be done and shaped in our democracy, without being finished” (Carl Friedrich v. Weizsäcker, 1965).
The foundation awards a yearly prize and medal to persons of high standing and organisations, which are groundbreaking in these areas of social commitment and the fostering of democracy.
This year the prize has been awarded to Oxfam and to FSFE. The FSFE press release is available (also in German, French, Italian, Dutch and Greek, thanks to our wonderful and dedicated translation teams). In these complicated days, it feels — to me, at least — a little weird to be side-by-side with an organization like Oxfam. They are helping rebuild Haiti. Providing direct support, like water and shelter; “helpin’ people with skin“, as Granny Weatherwax might say. The FSFE helps with skin, too, (for instance with licensing advice, workshops, and promotion booths at events), but for far more abstract goals. Which isn’t to say that Free Software does not help: some Free Software is really useful in a disaster.
It might feel a little weird in the current context, but it is an honor and a privilege.
KOffice on OpenSolaris
January 24th, 2010
Inge Wallin recently blogged about the portability of KOffice — spurred on, no doubt, by the success of the port to the Nokia n900 and to Haiku. So he listed GNU/Linux, Mac OSX, Windows, FreeBSD (thanks, Inge, for checking), Haiku. That list is missing (Open)Solaris though, which as a UNIX flavor. ought to be a pretty simple target.
Of course, Solaris has been a primary target for OpenOffice for ages, so KOffice is a little late to the game on this particular platform. But I guess that’s my fault, since I’m one of the packagers for KDE4 on Solaris, and I hadn’t gotten around to it yet. So this weekend I spent a little under two hours hammering together a specfile (RPM-style) for koffice and getting the whole darn thing to build. Screenshot of KWord in action as proof. I tried KPresenter as well, but that crashed on changing the list style, so I didn’t think that was a good demonstration.
Of course, no port is without its patches, so here they are:
- constness — this patch matches the constness of parameters in definitions with those in declarations, so that int foo(const int) is defined with int foo(const int i) and not as int foo(int i). Now, since the last time constness-matching blew up, I’ve learned that this is really a bug in Sun’s compiler, because constness is not supposed to be mangled into the name. However, I’ll claim that code neatness demands that they match, anyway.
- double — some math functions like sqrt() can take both float or double, and Sun’s compiler doesn’t just pick one when you pass it integer parameters, so we need some explicit disambiguation. This is common all across the KDE codebase.
- NaN — the KOffice code uses val != NAN which uses the gcc-specific NAN #define; probably !isnan() is better.
- math — the header file math.h in Solaris uses the identifier “exception”, which becomes ambiguous in the context of the STL, so it needs to be included earlier, rather than later. This patch bungs in math.h as the first include in a number of C++ files — not necessarily something to merge upstream, because it’s mostly working around a bug elsewhere.
- stupid — yes, this is a stupid patch. I can’t convince Sun Studio that 8.5 * 1440 is an secretly an integer constant and that it shouldn’t complain about a bunch of initializers in the MS Word import filters. I’m not really sure what’s going on here, it’s something special about static const initializers of class variables, since doing the same in a non-class context works just fine.
So there you have it. Five tiny patches for a codebase of a little over 600000 lines of code. Nice. Tip of the hat (I have lots of hats, but all of them are baseball caps and I don’t have a decent Stetson) to Inge and the KOffice folks for a nice portable office suite. You may be troubled, but your code is good.
On freeing a project
January 23rd, 2010
I saw over on LWN.net (a very valuable resource for technology articles, well worth your $5), that a RAW picture editor called RawTherapee had been re-licensed from Freeware (gratis, but proprietary) to Free Software, under the GPLv3. The application is something I won’t use myself, and I can’t even manage the comparison to DigiKam (which I don’t use either). Technologically it looks quite cool, from the cmake-based build (yes, having a build system that supports cross-platform development effectively can be a wonderful thing) to the (comprehensive and multi-lingual) ODT manuals. I miss the nice KDE reporting of missing dependencies summarized in a list at the end, though. Something to contribute, over there.
The author, Gábor Horváth, states some interesting reasons for freeing the whole thing; I personally think the “family” argument is the strongest one, and it’s important to spend time with your (small) kids. Of course, at 10 months you can still code with the baby on your tummy. The other motivations are largely a matter of growing the community of developers and contributors in order to get better coverage of all the aspects of development (GUI, documentation, algorithms, translations) and smoother development by not depending on a single individual. All good reasons, to be sure. Whether these are good motivations to choose GPLv3, I don’t know (over some other copyleft license that encourages community building).
I’m going to admit a very non-charitable reason for looking more closely at RawTherapee: I was intending to bitch and moan about the state of the source code and the licensing and then point out how it should be done, by gorm. Instead, I can point to RawTherapee (as of SVN revision 31) as an example of a Free Software project done (pretty) well:
- (good) Clear license choice; although the license is in a file called “copying” instead of the canonical “COPYING”, it’s the complete text of the GPLv3.
- (good) Every file has a complete copyright header, including sufficient license text and an identifiable copyright holder. As the project grows, it will be interesting to see if it manages to keep a good grip on its contributors — that is, if everyone who adds a copyrightable contribution to the code actually adds a meaningful copyright holder line to the file headers. This would be, roughly, clause 5a of the GPLv3.
- (bad) The distribution from SVN includes a few pre-compiled libraries (jpeg, lcms, png, tiff, z) for windows and accompanying header files. Of that lot, jpeg doesn’t mention its license terms in the file header (but refers to a README that has been lost). I hunted down the relevant source distribution, and the README contains a license that stipulates that README must be included with the distribution if any source is distributed. Right, minor point, but something to clean up in a next release — although I’d like to think that it also illustrates that the license in the jpeglib header should be clearer. lcms is MIT licensed, the png license doesn’t seem to allow binary distribution at all — the permissions it gives are for source. On the other hand, the use of the code as a component in a product is encouraged — that suggests binary use is ok, but it’s not exactly explicit. Tiff is MIT again (when I say “MIT” I mean “Simple permissive 1-clause, possibly with a restriction on endorsement”) and libz is also permissive. So just one minor blot.
Anyway, a tip of the hat to Gábor for doing it right (and doing the right thing) and I wish him lots of success with his project.