It wasn't really a flamewar, and isn't of the same importance as the release coordinator quitting.
One of the more abrasive characters in KDE circles is called Mosfet (real name Daniel Duley I think). He implemented much of the KDE 2 theming/widget styling support, as well as the image management software Pixie. He has a habit of finishing big batches of code about a week after the KDE code freeze dates, and then stamping his feet until he's allowed to check the code into CVS.
A while back, people were trying to reorganise the kdebase package, and split some of the less essential sections into two new packages: kdeaddons and kdeartwork. Mosfet had about seven or eight window border themes in kdebase, and they wanted him to move the less used ones into the addons package. He refused. Similarly for the less used widget styles. Then Mosfet developed a new widget style, which they wouldn't let him commit to CVS because it was past the feature freeze deadline for 2.2.
Mosfet, upset at this, decided to remove almost all his themes & styles from the KDE CVS, including the new default widget style. When this didn't get the reaction he expected, he removed Pixie from CVS as well (this is fair enough - there's no real reason for Pixie to be in the kdebase package anyhow: it's a distinct complete application).
A few days later, and with a cooler head, he's moved the styles KDE actually needs back into their CVS server. The rest will either go back in the future, or be a seperate download (i.e. from kde.themes.org).
And that's about the most exciting dissent to happen within KDE development for the last 18 months:)
Re:How to get attention to Ogg Vorbis
on
MP3Pro Released
·
· Score: 2
How about
this one?:)
It's currently up to version 1.05, and it'll be in the main Winamp distribution the next time they release an upgrade to Winamp.
Re:Why it's so small and why you want to avoid it
on
MP3Pro Released
·
· Score: 1
You may want to hang on just a little longer, for the 1.0 release candidate -- they're just finished implementing all the features in the Vorbis 1 spec (cascading, etc.) which the beta4 decoder/encoder doesn't understand. beta4 Ogg Vorbis files will still be decoded fine, but 1.0 files may not play correctly under the beta4 decoder (but hey, that's why it's beta:).
This is good news - beta4 already compresses better than standard MP3, and 1.0 should only improve this.
The problem with MP3 isn't content protection -- the problem is the licensing. There is no need to pay to get a license to *decode* MP3, but there is to *encode*. Even if you write a free encoder, you are supposed to pay for a license.
Given this, and despite what they say on their website, Lame and all other free MP3 encoders are unlicensed, illegal software in the USA.
The only reason this hasn't been pursued by Fraunhofer is the bad press it would generate -- but never rely on a company not to change its mind. One parallel: GIF and the UNISYS patent.
Asking the average teenager on the street anything and expecting them to tell you anything moderately intelligent is a triumph of hope over experience.
All a new file format needs to be successful is for the people that pirate movies and CDs to start using it. One of the triggers for the success of.mp3 was the warez scene, and the same could happen for.ogg.
The alternative route is for the file format to mysteriously become the default on the operating system that all these technically semi-literate people use (as Microsoft will be trying to do with.wma).
The real advantage of.ogg over *all* the other recent lossy perceptual encoders is that it has *no* patenting issues, *no* licensing issues, and has equal or better performance than the current market leader. All it needs now is for some people with a marketing budget to pick it up:)
What about a four letter extension?.html
You'll find that it's quite popular.
But that's beside the point.
Most people don't look at file formats *at all* on Windows these days - they have pretty icons which shows them what the file type is.
But that's beside the point as well.
The Ogg file format extension is.OGG, and Ogg is a general packet based streaming file format. The Vorbis audio format is the first user of this system. So, in a way,.OGG files are intended to be competitors to QuickTime.MOV, which also encompasses a wide variety of codecs.
Re:Bleeding edge compatibility
on
MP3Pro Released
·
· Score: 3
The best Windows based ripper/encoder is
CDex. It's small, light, fast, fully featured and free. Hard to think of a better combination of features (yes, it's also open source, but that's beside the point).
For Linux, use cdparanoia + lame or oggenc, or one of the many good frontends - Grip comes to mind. Additionally, the upcoming KDE 2.2's builtin audiocd IOSlave will allow you to rip CDs very easily (though not yet write them).
You 'gave up after 5 albums' - why? You can use the computer for other things while you're ripping a CD, you know:). Combine ripping and encoding, and it'll probably take about 30 minutes to fully process a CD, but there's no reason why this shouldn't go on in the background.
180 or 200? That would make it a *very* exclusive group:)
If course, I'm not sure if we're just more stupid over here, but in the UK the entry requirement is around 150, and that's fudged a lot of the time. They give you 3 IQ tests, and if any of them are over the required amount, they say 'sure, you can join us, just pay £x a month'.
The real intelligence test in the Mensa application process is realising that you don't need to join a society to show that you are clever.
(offtopic)
Python doesn't force you to use TABs. It *does*, however, force you to indent consistently. That's it. You don't have to use 8 spaces, or even a set number of spaces - as long as the indentation is consistent within a block.
Many people who program in Python dislike TABs, and set their editors to automatically replace then with a set number of spaces. Better still, use a decent editor like VI or EMACS, which understand Python and let you navigate around it intelligently (just like I wouldn't want to program Lisp without the Emacs Lisp editing mode).
When you're writing C, you indent AND use braces -- with Python you only indent. Less potential for confusion.
(/offtopic)
My university (Warwick University, UK), have switched to using Java as their introductory programming language as well. I may be biased, but it seems a very bad idea for a number of reasons. Principally, Java is a complicated multi paradigm language -- and when you're teaching, you want to be able to get the ideas across cleanly, without pages of 'just write all this, don't worry about what it means' which you'd have to use in Java.
For structured programming, use Pascal (it's what it was designed for, and it fulfills its purpose incredibly well). For functional programming, use Lisp or fashionable modern alternatives like Haskell or ML. For object orientation, Python is a good choice, or Smalltalk if you want to be more hardline. (a few years ago, I might have mentioned Prolog, but that seems to have fallen by the way-side along with expert systems and hard AI).
Java, along with C and C++, are very bad as *teaching* languages. Once you know the basics, however, they are great languages to learn precisely because they are multiparadigm, and don't force you into a particular programming style.
I wonder how many "I wonder how many "I wonder how many "I wonder how many "I wonder how many "I wonder how many "I wonder how many "I wonder how many "I wonder how many "all your tech support is belong to us!" posts we're gonna see here..." we're gonna see here..." we're gonna see here..." we're gonna see here..." we're gonna see here..." we're gonna see here..." we're gonna see here..." posts we're gonna see here..." we're gonna see here...
You may have noticed that I tested all the links pointed out in both 2.1 and a recent (Sunday/Monday) CVS copy of 2.2alpha. As you mentioned, 3 of the links are incorrect in 2.1 but work in 2.2.
Why does it matter that KHTML doesn't render CSS1 correctly? Because on the
Konqueror website it claims to render CSS1 perfectly apart from 3 attributes. It's not being unfair or nasty to point out that this claim is currently wrong. I'm currently submitting all the links he noticed to the
KDE bugs database
Konqueror and KHTML are a perfect advertisment for the open source development model. As you say, it works almost perfectly in 'the real world'... which is why we have to look at these marginal cases of 'features no-one uses' to find problems:). Imagine what the KDE people could have done with 10% of the money that's been thrown at Mozilla these last three years... (mind you, more money can sometimes be a hindrance - look at the Nautilus fiasco).
No argument here - they're claiming to be able to do more than they actualy can, at the moment. Sorry if I jumped on you a little bit -- I'm really impressed with how quicked Konqeuror has developed, and get annoyed when people who don't know better criticise. You, however, obviously DO know what you're talking about (which makes a change on Slashdot:). Considering how long they've been developing Konqueror, and how few people they have full (or even half) time, the fact they we stand here just moaning about bits of CSS is a huge credit to them.
Getting back back to the discussion topic, I'm really happy that Mozilla is finally coming along, because it IS the most standards compliant browser. I'm just worried that, because of the length of time it's taken to get to a releasable state, no one in the outside world will care when it's finally available. Like many people, when Mozilla started I used to check up on it all the time, and try all the nightlies. But months went by with no visible implovements, and I found myself spending more and more of my browsing time with Konqueror. I don't see how Mozilla will adequately compete with Konqueror on KDE UNIX boxes and IE on Windows boxes, both deeply integrated into their respective environments.
"If anyone is really interested, I'll post a list of sites that Opera and Konqueror foul up that Mozilla get right." vs.
"You missed my point. I am not interested in developing Konqueror. "
I'm sorry I made you unhappy by responding in a calm and measured way. I would like ALL the open source browsers to be standards compliant, and because of this, thanks for the links. I'm currently working through them.
Let's look at them in order:
Small caps. Not rendered acceptably in 2.1 or 2.2alpha. Bug report will be submitted.
Backgrounds. Incorrect in 2.1, working in 2.2alpha.
Word spacing. Incorrect in 2.1 and 2.2alpha.
Float/clear. Incorrect in 2.1 and 2.2alpha.
Whitespace handling. Incorrect in 2.1 and 2.2alpha.
CSS borders. Incorrect in 2.1, working in 2.2alpha.
Table rendering. Incorrect in 2.1 and 2.2alpha.
Robust parser. Incorrect in 2.1 and 2.2alpha.
Styling elements. Partially correct in 2.1 and 2.2alpha.
Background image positioning. Incorrect in 2.1, working in 2.2alpha.
Z-index. Incorrect in 2.1 and 2.2alpha.
site for sick writers. Looked okay to me:)
stupid and cool. Incorrect in 2.1 and 2.2alpha (probably a Javascript issue).
overflow hidden. Buggy in 2.1 and 2.2alpha.
Thanks for your input. I would however say that Konqueror seems quite a bit better than Netscape 4.x ever was:).
Remember that it is, sadly, not the standards organisations that define how people write their web pages. Regardless of how Konqueror and Mozilla render pages, if they render them differently from Internet Explorer then they are wrong and IE is right, in the only sense that most end users will care about.
Because sometimes the fixes REQUIRE a rewrite. And for the obnoxious and ignorant individuals who get righteous and say "they should have done it right the first time": then what the hell do we need version numbers FOR?
I talk about version numbers later.
On the rewrite point... yes, sometimes you get so far and realise that you were wrong, and that you need to throw code out and rewrite. Something similar happened to KDE in the development process for KDE 2, when they decided (rightly or wrongly) that they couldn't do what they wanted with CORBA, and had to rewrite all their mechanisms using DCOP. But Mozilla has had *so many* API changes and rewrites, *so many* roadmap delays and plan alterations, that the incredible amount of expectation and mindshare it had even up to a year ago has just fizzled away.
I try to be rational and see all the problems they had, and look forward to the excellent end result. But the process has been so disappointing and disenchanting. Sometimes I get a little too heated when people talk about it as an unmitigated success:)
Other people can feel free to grab one of the milestones, and do a search and replace for the version number to make it "1.0".
Very true:). Sometimes I think that it's a real pity that we only use 1 dimension for version numbers. This imposes a linear view -- that things are always getting better in time, whereas sometimes you have to break things to make them better in the end. Something nice and multidimensional would fit the development process better ('this release adds 1 to the criteria 5 axis, but subtracts 3 from the criteria 8 axis. Subsequent releases in this quarter will focus on criteria 8':). You could map to a scalar by taking the modulus, and sometimes this modulus would decrease.
Sorry, but there is no way you can hold up the Mozilla development process as a standard of excellence.
There are two seperate issues here, which are getting confused by some people. The first is the one you mentioned: that.0 releases should be fuctionally complete and bug free. This is a laudable aim, but not always achievable, particularly in the context of an enthusiast led, non full-time development workforce. Sometimes, as with the Linux 2.4.0 release, you *need* to release something as.0 in order to get your workforce to rally round and start debugging:). There's an apt quote about 'herding cats' about this on the tip of my tongue...
Of course, in a commercial environment forcing people to debug in such a drastic fashion isn't necessary (in some utopial ideal world:) ). Given that, until recently, almost all the coding on Mozilla was being done by paid Netscape engineers, you would expect a degree of planning and management to be evident. So, the second issue is the Mozilla development process itself. Mozilla seems to have turned in to a 'developers playground', initially overdesigned and poorly implemented, and subsequently rewritten ad infinitum.
It's very possible that we could have had a decent, workable Mozilla 1.0 a year ago, had the developers focused on reasonable aims. Instead they wanted to write the browser-to-end-all-browsers at their very first try, and this interminable delay is the result.
The annoying thing is that this failure of a commercial company, developing in a commercial bloated over designed fashion, has been held up as an example of the failures of Open Source development.
Exactly. Some of the Mozilla subsystems are in their third rewrite, and we've not had a version 1 yet! I don't see why they can't just say "right. no more API changes, no more rewrites. let's spend the next 3 months making everything stable, and release 1.0". Then they can release a 1.1 6 months later with whatever enhancements they want. They shouldn't be shooting for perfection with their first stable release.
Look at KDE (or XFree86 or the Linux Kernel). Eventually, they got to a point in the development of their 2.0 (resp. 4.0 resp. 2.4) release where they realised that they could fritter their lives away trying for perfection. With KDE, they made a 2.0 release that wasn't fully satisfactory, but it gave them a base to get feedback and build on that they just wouldn't have had if they hadn't released it -- and 2.1.1/2 is the stable result. The same goes for the other two projects mentioned -- eventually you just have to stop adding features.
The incredible thing about Mozilla is how incredibly badly the project has been managed. KDE have written an entire desktop environment since Mozilla was announced... governments have fallen... continents have crumbled... It's a perfect example of how *not* to manage a large complicated project.
If you know of any websites which Konqueror doesn't correctly display, then submit a bugreport to them (http://bugs.kde.org). It already in 2.1.1 copes with a much wider range of poorly coded websites than 2.0 did. That would take about the same time as posting a list to here, and would be much more useful to the developers. Believe it or not, the developers don't sit back all day, cackling at the fact that their programs doesn't let you view pages correctly -- they can't fix problems until they have test cases, and it sounds like you have found some.
Post them here if you like, and I'll verify that they don't work on the most recent CVS development snapshot. I'll even submit the bug report for you if you want.
It's not just the Internet that Microsoft might control.
Take a look at
this piece in The Register. Basically, Microsoft have implemented a site for the UK government called
Government Gateway, which will enable you to use your computer to electronically perform a lot of tasks which previously needed lots of paper work (like Tax Self Assessment). However, if you go to the Gateway you find that they have very restrictive checks on the browser you are using -- and they won't let you use some of the areas which use a digital certificate unless you are using Internet Explorer 5+.
You can still use it if you fake the UserAgent string, but this sort of behaviour from a website is at best crude, and at worst deliberately targeting non-Microsoft OSes.
The problem is that 70% should *not* mean 'tolerable', it should mean 'good, but not excellent'. 50% should mean 'average'. The excellent UK magazine
EDGE follows this philosophy. These are the approximate meanings of the EDGE game ratings (taken from their review section):
disastrous
appalling
severely flawed
disappointing
average
competent
distinguished
excellent
astounding
revolutionary
I think they have only ever given 3 games 10/10 - two of those were Gran Turismo and Mario 64. This honesty is one of the main reasons I read the magazine (that and the impression that it's written for adults, and game designers, rather than 12 year olds). Certainly makes a change from Spectrum User, when you could tell a game was good if it got more than 95%:).
Re:Possibly for somethings, not all though.
on
Qt for Mac
·
· Score: 2
If people want to release their coding endeavors as shareware, that is their right.
You are right. I'm not a raving free software bigot. I've nothing against the concept of shareware -- the classic example, as mentioned in a post below, being the early versions of Paint Shop Pro. A decent product at a decent price. What I dislike is the proliferation in the Windows world of small tools which have very little functionality, but which have a grossly oversized pricetag attached to them. The only thing this produces is a flourishing warez culture. For example, how many people do you know that use WinZip and have actually paid for it?
I really believe the state of shareware on Windows has been a significant cause for the endemic illegal copying of software on that platform. Windows users get used to the fact that simple tools will be cripple- or nag- ware until they (or their tech savvy friend) downloads a crack for the product. This attitude then bleeds upwards, infecting the market for full commercial software such as Photoshop and Microsoft Office.
Why do you believe shareware is a good thing?
Re:Possibly for somethings, not all though.
on
Qt for Mac
·
· Score: 2
This is a good thing. The worst aspect of the Windows software scene is the proliferation of shareware apps in areas where the code should be made freely available. By putting up a modest barrier to entry, Trolltach are actively discoraging small scale shareware, and I applaud them for it.
Re:Who said Gnome is dead too? Why GNOME is doomed
on
Eazel Come, Eazel Go?
·
· Score: 1
(this is a response to an AC for those of you browsing at >=+1).
You're being a little unfair in a couple of different ways.
The state of Gnome development at the moment is a little like KDE's was between the releases of 1.1.2 and 2.0. They decided to completely restructure the design of the environment. This is very interesting as a core development, but means very little visible progress from the user perspective. At the moment, Gnome are nailing down exactly what they want in 2.0 -- and, as always, documentation lags behind (the perrenial problem of software development). The gnome.org website could be improved, but don't judge an entire project by the state of their website!
On your second point -- it is not necessarily stupid to write an object based system in C. The advantage of C is that you get full and complete control over your system. The disadvantage, of course, is that you get full and complete control over your system:) C++ also, is not the ideal language to write an object based system in. It's the PL/1 or ADA of the modern world (on an offtopic note, I have the ANSI reference manuals for ADA-1983, PL/1-1983 and FORTRAN-1978 on my home bookshelf next to me. All good for a browse when bored. ADA doesn't seem that bad - why does everyone hate it?).
Finally, don't be 'pro-GNOME' or 'pro-KDE'. (or for that matter 'pro-XFCE' - one man's attempt to write a better-CDE-than-CDE). Support all of them.
(we don't seem to have had a decent KDE/GNOME flamewar for a little while -- guess it was about time for another one:).
It's very odd when you can get so worked up about a couple of buttons. They're all just pixmaps, you know -- if you really wanted, I'm sure you could change them over to look very similar to the Gnome ones in a couple of days. They're always on the lookout for graphic artists.
Of course, there are problems with the themability of KDE. It's not that KDE can't be themed as much as Gnome (it is just as flexible) -- but that theming Gnome is much easier for the end user. This is being worked on. Bother usability and design are more important than looks. That said, personally, I think KDE 2 looks *so* much better than KDE 1 (and 2.1 more polished than 2) that I don't really mind them taking their time about it.
You're right about corner panels -- I always thought you *could* convert the thing to a corner panel, but I guess not (there is Kasbar though). Submit a wishlist bug report and see what they say.
Also, I think you are overstating a little. you don't *hate* things, you *dislike* them. If you were tied to a computer and forced to use KDE, you could survive. Lighten up a little.
The world is a better place for having both GNOME and KDE in it. (although I still just use FVWM + xterms at work:).
Re:Good riddance to yet another bad business model
on
Eazel Come, Eazel Go?
·
· Score: 1
I'm not sure if you actually have any experience with commercial or shareware development, but £1000 for an unlimited commercial use license is actually quite cheap. Look up how much Motif used to cost way back when.
In fact, you could argue that QT being GPL not LGPL is *good* for the open source community. It ensures that there is a small barrier to entry for people thinking of making shareware programs. The last thing we need is to get the same shareware scene in UNIX as has existed in Windows.
Slashdot Mirror
It wasn't really a flamewar, and isn't of the same importance as the release coordinator quitting.
:)
One of the more abrasive characters in KDE circles is called Mosfet (real name Daniel Duley I think). He implemented much of the KDE 2 theming/widget styling support, as well as the image management software Pixie. He has a habit of finishing big batches of code about a week after the KDE code freeze dates, and then stamping his feet until he's allowed to check the code into CVS.
A while back, people were trying to reorganise the kdebase package, and split some of the less essential sections into two new packages: kdeaddons and kdeartwork. Mosfet had about seven or eight window border themes in kdebase, and they wanted him to move the less used ones into the addons package. He refused. Similarly for the less used widget styles. Then Mosfet developed a new widget style, which they wouldn't let him commit to CVS because it was past the feature freeze deadline for 2.2.
Mosfet, upset at this, decided to remove almost all his themes & styles from the KDE CVS, including the new default widget style. When this didn't get the reaction he expected, he removed Pixie from CVS as well (this is fair enough - there's no real reason for Pixie to be in the kdebase package anyhow: it's a distinct complete application).
A few days later, and with a cooler head, he's moved the styles KDE actually needs back into their CVS server. The rest will either go back in the future, or be a seperate download (i.e. from kde.themes.org).
And that's about the most exciting dissent to happen within KDE development for the last 18 months
How about this one? :)
It's currently up to version 1.05, and it'll be in the main Winamp distribution the next time they release an upgrade to Winamp.
You may want to hang on just a little longer, for the 1.0 release candidate -- they're just finished implementing all the features in the Vorbis 1 spec (cascading, etc.) which the beta4 decoder/encoder doesn't understand. beta4 Ogg Vorbis files will still be decoded fine, but 1.0 files may not play correctly under the beta4 decoder (but hey, that's why it's beta :).
This is good news - beta4 already compresses better than standard MP3, and 1.0 should only improve this.
The problem with MP3 isn't content protection -- the problem is the licensing. There is no need to pay to get a license to *decode* MP3, but there is to *encode*. Even if you write a free encoder, you are supposed to pay for a license.
Given this, and despite what they say on their website, Lame and all other free MP3 encoders are unlicensed, illegal software in the USA.
The only reason this hasn't been pursued by Fraunhofer is the bad press it would generate -- but never rely on a company not to change its mind. One parallel: GIF and the UNISYS patent.
Asking the average teenager on the street anything and expecting them to tell you anything moderately intelligent is a triumph of hope over experience.
.mp3 was the warez scene, and the same could happen for .ogg.
.wma).
.ogg over *all* the other recent lossy perceptual encoders is that it has *no* patenting issues, *no* licensing issues, and has equal or better performance than the current market leader. All it needs now is for some people with a marketing budget to pick it up :)
All a new file format needs to be successful is for the people that pirate movies and CDs to start using it. One of the triggers for the success of
The alternative route is for the file format to mysteriously become the default on the operating system that all these technically semi-literate people use (as Microsoft will be trying to do with
The real advantage of
What about a four letter extension? .html
.OGG, and Ogg is a general packet based streaming file format. The Vorbis audio format is the first user of this system. So, in a way, .OGG files are intended to be competitors to QuickTime .MOV, which also encompasses a wide variety of codecs.
You'll find that it's quite popular.
But that's beside the point.
Most people don't look at file formats *at all* on Windows these days - they have pretty icons which shows them what the file type is.
But that's beside the point as well.
The Ogg file format extension is
For Linux, use cdparanoia + lame or oggenc, or one of the many good frontends - Grip comes to mind. Additionally, the upcoming KDE 2.2's builtin audiocd IOSlave will allow you to rip CDs very easily (though not yet write them).
You 'gave up after 5 albums' - why? You can use the computer for other things while you're ripping a CD, you know :). Combine ripping and encoding, and it'll probably take about 30 minutes to fully process a CD, but there's no reason why this shouldn't go on in the background.
180 or 200? That would make it a *very* exclusive group :)
If course, I'm not sure if we're just more stupid over here, but in the UK the entry requirement is around 150, and that's fudged a lot of the time. They give you 3 IQ tests, and if any of them are over the required amount, they say 'sure, you can join us, just pay £x a month'.
The real intelligence test in the Mensa application process is realising that you don't need to join a society to show that you are clever.
(offtopic)
Python doesn't force you to use TABs. It *does*, however, force you to indent consistently. That's it. You don't have to use 8 spaces, or even a set number of spaces - as long as the indentation is consistent within a block.
Many people who program in Python dislike TABs, and set their editors to automatically replace then with a set number of spaces. Better still, use a decent editor like VI or EMACS, which understand Python and let you navigate around it intelligently (just like I wouldn't want to program Lisp without the Emacs Lisp editing mode).
When you're writing C, you indent AND use braces -- with Python you only indent. Less potential for confusion.
(/offtopic)
My university (Warwick University, UK), have switched to using Java as their introductory programming language as well. I may be biased, but it seems a very bad idea for a number of reasons. Principally, Java is a complicated multi paradigm language -- and when you're teaching, you want to be able to get the ideas across cleanly, without pages of 'just write all this, don't worry about what it means' which you'd have to use in Java.
For structured programming, use Pascal (it's what it was designed for, and it fulfills its purpose incredibly well). For functional programming, use Lisp or fashionable modern alternatives like Haskell or ML. For object orientation, Python is a good choice, or Smalltalk if you want to be more hardline. (a few years ago, I might have mentioned Prolog, but that seems to have fallen by the way-side along with expert systems and hard AI).
Java, along with C and C++, are very bad as *teaching* languages. Once you know the basics, however, they are great languages to learn precisely because they are multiparadigm, and don't force you into a particular programming style.
I wonder how many "I wonder how many "I wonder how many "I wonder how many "I wonder how many "I wonder how many "I wonder how many "I wonder how many "I wonder how many "all your tech support is belong to us!" posts we're gonna see here..." we're gonna see here..." we're gonna see here..." we're gonna see here..." we're gonna see here..." we're gonna see here..." we're gonna see here..." posts we're gonna see here..." we're gonna see here...
Why does it matter that KHTML doesn't render CSS1 correctly? Because on the
Konqueror website it claims to render CSS1 perfectly apart from 3 attributes. It's not being unfair or nasty to point out that this claim is currently wrong. I'm currently submitting all the links he noticed to the
KDE bugs database
Konqueror and KHTML are a perfect advertisment for the open source development model. As you say, it works almost perfectly in 'the real world'... which is why we have to look at these marginal cases of 'features no-one uses' to find problems
No argument here - they're claiming to be able to do more than they actualy can, at the moment. Sorry if I jumped on you a little bit -- I'm really impressed with how quicked Konqeuror has developed, and get annoyed when people who don't know better criticise. You, however, obviously DO know what you're talking about (which makes a change on Slashdot :). Considering how long they've been developing Konqueror, and how few people they have full (or even half) time, the fact they we stand here just moaning about bits of CSS is a huge credit to them.
Getting back back to the discussion topic, I'm really happy that Mozilla is finally coming along, because it IS the most standards compliant browser. I'm just worried that, because of the length of time it's taken to get to a releasable state, no one in the outside world will care when it's finally available. Like many people, when Mozilla started I used to check up on it all the time, and try all the nightlies. But months went by with no visible implovements, and I found myself spending more and more of my browsing time with Konqueror. I don't see how Mozilla will adequately compete with Konqueror on KDE UNIX boxes and IE on Windows boxes, both deeply integrated into their respective environments.
"If anyone is really interested, I'll post a list of sites that Opera and Konqueror foul up that Mozilla get right." vs. "You missed my point. I am not interested in developing Konqueror. "
I'm sorry I made you unhappy by responding in a calm and measured way. I would like ALL the open source browsers to be standards compliant, and because of this, thanks for the links. I'm currently working through them.
Let's look at them in order:
- Small caps. Not rendered acceptably in 2.1 or 2.2alpha. Bug report will be submitted.
- Backgrounds. Incorrect in 2.1, working in 2.2alpha.
- Word spacing. Incorrect in 2.1 and 2.2alpha.
- Float/clear. Incorrect in 2.1 and 2.2alpha.
- Whitespace handling. Incorrect in 2.1 and 2.2alpha.
- CSS borders. Incorrect in 2.1, working in 2.2alpha.
- Table rendering. Incorrect in 2.1 and 2.2alpha.
- Robust parser. Incorrect in 2.1 and 2.2alpha.
- Styling elements. Partially correct in 2.1 and 2.2alpha.
- Background image positioning. Incorrect in 2.1, working in 2.2alpha.
- Z-index. Incorrect in 2.1 and 2.2alpha.
- site for sick writers. Looked okay to me
:)
- stupid and cool. Incorrect in 2.1 and 2.2alpha (probably a Javascript issue).
- overflow hidden. Buggy in 2.1 and 2.2alpha.
Thanks for your input. I would however say that Konqueror seems quite a bit better than Netscape 4.x ever wasRemember that it is, sadly, not the standards organisations that define how people write their web pages. Regardless of how Konqueror and Mozilla render pages, if they render them differently from Internet Explorer then they are wrong and IE is right, in the only sense that most end users will care about.
I talk about version numbers later. On the rewrite point... yes, sometimes you get so far and realise that you were wrong, and that you need to throw code out and rewrite. Something similar happened to KDE in the development process for KDE 2, when they decided (rightly or wrongly) that they couldn't do what they wanted with CORBA, and had to rewrite all their mechanisms using DCOP. But Mozilla has had *so many* API changes and rewrites, *so many* roadmap delays and plan alterations, that the incredible amount of expectation and mindshare it had even up to a year ago has just fizzled away.
I try to be rational and see all the problems they had, and look forward to the excellent end result. But the process has been so disappointing and disenchanting. Sometimes I get a little too heated when people talk about it as an unmitigated success :)
Other people can feel free to grab one of the milestones, and do a search and replace for the version number to make it "1.0".
Very true :). Sometimes I think that it's a real pity that we only use 1 dimension for version numbers. This imposes a linear view -- that things are always getting better in time, whereas sometimes you have to break things to make them better in the end. Something nice and multidimensional would fit the development process better ('this release adds 1 to the criteria 5 axis, but subtracts 3 from the criteria 8 axis. Subsequent releases in this quarter will focus on criteria 8' :). You could map to a scalar by taking the modulus, and sometimes this modulus would decrease.
Sorry, but there is no way you can hold up the Mozilla development process as a standard of excellence.
.0 releases should be fuctionally complete and bug free. This is a laudable aim, but not always achievable, particularly in the context of an enthusiast led, non full-time development workforce. Sometimes, as with the Linux 2.4.0 release, you *need* to release something as .0 in order to get your workforce to rally round and start debugging :). There's an apt quote about 'herding cats' about this on the tip of my tongue...
:) ). Given that, until recently, almost all the coding on Mozilla was being done by paid Netscape engineers, you would expect a degree of planning and management to be evident. So, the second issue is the Mozilla development process itself. Mozilla seems to have turned in to a 'developers playground', initially overdesigned and poorly implemented, and subsequently rewritten ad infinitum.
There are two seperate issues here, which are getting confused by some people. The first is the one you mentioned: that
Of course, in a commercial environment forcing people to debug in such a drastic fashion isn't necessary (in some utopial ideal world
It's very possible that we could have had a decent, workable Mozilla 1.0 a year ago, had the developers focused on reasonable aims. Instead they wanted to write the browser-to-end-all-browsers at their very first try, and this interminable delay is the result.
The annoying thing is that this failure of a commercial company, developing in a commercial bloated over designed fashion, has been held up as an example of the failures of Open Source development.
Exactly. Some of the Mozilla subsystems are in their third rewrite, and we've not had a version 1 yet! I don't see why they can't just say "right. no more API changes, no more rewrites. let's spend the next 3 months making everything stable, and release 1.0". Then they can release a 1.1 6 months later with whatever enhancements they want. They shouldn't be shooting for perfection with their first stable release.
Look at KDE (or XFree86 or the Linux Kernel). Eventually, they got to a point in the development of their 2.0 (resp. 4.0 resp. 2.4) release where they realised that they could fritter their lives away trying for perfection. With KDE, they made a 2.0 release that wasn't fully satisfactory, but it gave them a base to get feedback and build on that they just wouldn't have had if they hadn't released it -- and 2.1.1/2 is the stable result. The same goes for the other two projects mentioned -- eventually you just have to stop adding features.
The incredible thing about Mozilla is how incredibly badly the project has been managed. KDE have written an entire desktop environment since Mozilla was announced... governments have fallen... continents have crumbled... It's a perfect example of how *not* to manage a large complicated project.
If you know of any websites which Konqueror doesn't correctly display, then submit a bugreport to them (http://bugs.kde.org). It already in 2.1.1 copes with a much wider range of poorly coded websites than 2.0 did. That would take about the same time as posting a list to here, and would be much more useful to the developers. Believe it or not, the developers don't sit back all day, cackling at the fact that their programs doesn't let you view pages correctly -- they can't fix problems until they have test cases, and it sounds like you have found some.
Post them here if you like, and I'll verify that they don't work on the most recent CVS development snapshot. I'll even submit the bug report for you if you want.
Take a look at this piece in The Register. Basically, Microsoft have implemented a site for the UK government called Government Gateway, which will enable you to use your computer to electronically perform a lot of tasks which previously needed lots of paper work (like Tax Self Assessment). However, if you go to the Gateway you find that they have very restrictive checks on the browser you are using -- and they won't let you use some of the areas which use a digital certificate unless you are using Internet Explorer 5+.
You can still use it if you fake the UserAgent string, but this sort of behaviour from a website is at best crude, and at worst deliberately targeting non-Microsoft OSes.
No need to worry. Going on past evidence, I expect release 1.0 of Mozilla to arrive some time around 2005...
I think they have only ever given 3 games 10/10 - two of those were Gran Turismo and Mario 64. This honesty is one of the main reasons I read the magazine (that and the impression that it's written for adults, and game designers, rather than 12 year olds). Certainly makes a change from Spectrum User, when you could tell a game was good if it got more than 95% :).
You are right. I'm not a raving free software bigot. I've nothing against the concept of shareware -- the classic example, as mentioned in a post below, being the early versions of Paint Shop Pro. A decent product at a decent price. What I dislike is the proliferation in the Windows world of small tools which have very little functionality, but which have a grossly oversized pricetag attached to them. The only thing this produces is a flourishing warez culture. For example, how many people do you know that use WinZip and have actually paid for it?
I really believe the state of shareware on Windows has been a significant cause for the endemic illegal copying of software on that platform. Windows users get used to the fact that simple tools will be cripple- or nag- ware until they (or their tech savvy friend) downloads a crack for the product. This attitude then bleeds upwards, infecting the market for full commercial software such as Photoshop and Microsoft Office.
Why do you believe shareware is a good thing?
This is a good thing. The worst aspect of the Windows software scene is the proliferation of shareware apps in areas where the code should be made freely available. By putting up a modest barrier to entry, Trolltach are actively discoraging small scale shareware, and I applaud them for it.
(this is a response to an AC for those of you browsing at >=+1).
:) C++ also, is not the ideal language to write an object based system in. It's the PL/1 or ADA of the modern world (on an offtopic note, I have the ANSI reference manuals for ADA-1983, PL/1-1983 and FORTRAN-1978 on my home bookshelf next to me. All good for a browse when bored. ADA doesn't seem that bad - why does everyone hate it?).
You're being a little unfair in a couple of different ways.
The state of Gnome development at the moment is a little like KDE's was between the releases of 1.1.2 and 2.0. They decided to completely restructure the design of the environment. This is very interesting as a core development, but means very little visible progress from the user perspective. At the moment, Gnome are nailing down exactly what they want in 2.0 -- and, as always, documentation lags behind (the perrenial problem of software development). The gnome.org website could be improved, but don't judge an entire project by the state of their website!
On your second point -- it is not necessarily stupid to write an object based system in C. The advantage of C is that you get full and complete control over your system. The disadvantage, of course, is that you get full and complete control over your system
Finally, don't be 'pro-GNOME' or 'pro-KDE'. (or for that matter 'pro-XFCE' - one man's attempt to write a better-CDE-than-CDE). Support all of them.
(we don't seem to have had a decent KDE/GNOME flamewar for a little while -- guess it was about time for another one :).
:).
It's very odd when you can get so worked up about a couple of buttons. They're all just pixmaps, you know -- if you really wanted, I'm sure you could change them over to look very similar to the Gnome ones in a couple of days. They're always on the lookout for graphic artists.
Of course, there are problems with the themability of KDE. It's not that KDE can't be themed as much as Gnome (it is just as flexible) -- but that theming Gnome is much easier for the end user. This is being worked on. Bother usability and design are more important than looks. That said, personally, I think KDE 2 looks *so* much better than KDE 1 (and 2.1 more polished than 2) that I don't really mind them taking their time about it.
You're right about corner panels -- I always thought you *could* convert the thing to a corner panel, but I guess not (there is Kasbar though). Submit a wishlist bug report and see what they say.
Also, I think you are overstating a little. you don't *hate* things, you *dislike* them. If you were tied to a computer and forced to use KDE, you could survive. Lighten up a little.
The world is a better place for having both GNOME and KDE in it. (although I still just use FVWM + xterms at work
I'm not sure if you actually have any experience with commercial or shareware development, but £1000 for an unlimited commercial use license is actually quite cheap. Look up how much Motif used to cost way back when.
In fact, you could argue that QT being GPL not LGPL is *good* for the open source community. It ensures that there is a small barrier to entry for people thinking of making shareware programs. The last thing we need is to get the same shareware scene in UNIX as has existed in Windows.