OSNews

It's been years since I used GTK+ and the Gnome API, so it won't be 5 minutes for me. It will be 5 hours. It's 5 minutes for the guy who owns the app though.

The change seems simple, but the last comment reveals a bigger issue. In order to implement your good change, GNOME would need to create a new theme element. You can't just hard-code the background as grey since if you use a theme with grey letters, nothing will be readable.

That being said, there is another approach -- reuse existing themed elements. For instance, turn each section into into a tab:

-------------
Section 1 |
-----------------------------------------------
Items in section 1

-------------
Section 2 |
-----------------------------------------------
Items in section 2

At the moment, that page looks like it's just a huge custom control that draws bold text for the sections and then draws the section links and you can click on.

Drawing the darkened banner on the back is not that hard, even if you are worried about the usability repercussions. There must be the GTK equivalent to the Win32 function GetSysColor which allows you to retrieve the various colors associated with the current theme. One possibility in this scenario is to make the background of the banner use the color of controls (like buttons) and have the text colored the same as the theme's text color.

That way, if you can read the text on a pushbutton on a particular theme, you should have no issues with the banner. If you can't read the text, you're probably going to have issues reading text on controls anyway so go shoot your theme designer.

That makes no sense what-so-ever.

I don't see why it needs to be a theme colour, there is a very simple solution for it, just have a semi-transparent gray overlay when the background is light and semi-transparent white overlay when the background is dark.

This is a simple solution that will work with any theme, probably not 5 minutes but I doubt it would be an hours work.

Stop making excuses, the developer is obviously lazy

Sounds awesome, get coding since you seem to have it all figured out already!

I didn't say *I* wasn't lazy

I hope that was supposed to be a joke? Either that or you have no idea about what you're talking about.

Is that supposed to be a troll post? You give no argument at all, are you kidding me?

It's not hard to calculate the brightness of a colour and I assume GTK has some sort of alpha-blending capabilities for canvas/graphics context/whatever they call it (even if it hasn't; a function can be implemented within 20 lines of code).

Next time it would be nice to have a little more insight into why you "think" something is not feasible.

Five hours seems like a tiny investment for something that you have apparently been harassing the gnome developers about for over a year :-)

Would be just as easy to not have a good moan about it too. (or maybe they might, just might accept a patch. You won't know unless you try it.)

Am I right in thinking that the original dig was slightly trollish?

No, you're wrong. I wasn't slightly trollish, it was big time trollish.

But then again, anyone who's been around this site for some time wouldn't expect anything less from Eugenia.

Yeah, typical answer of a person that does not want to provide a patch or any help, say it is not going to happen, even though they don't know if it is going to. Please, next time, file a bug report and submit a patch if you can. People just sit there waiting for developers to do everything, there is one simple factor, open source developers are most of the time voluntary workers. Windows is hacked on by a lot of people that get paid a lot of money, yet they need vendors to make drivers for them. And they make an OS that doesn't just cut it for me and is too buggy.

But well, people are being infected by the Linux haters blog syndrome, a person that does not contribute and just whines, how mature of him.

Edited 2008-07-06 04:03 UTC

Translation: why bother trying to fix a problem when you can just whinge about it endlessly?

It would not need to set in your package. If you want something changed, build it yourself.

Does that 5 hours for you include all the requisite QA that really needs to be done to verify that it doesn't screw up more than it supposedly fixes?

If you have serious experience in development with complex projects, you'd be more aware that those 5 minute changes often have much greater repercussions than you'd think, and you need to verify that you haven't forgotten something before you submit it into the great wide open. Something that can be added in 5 minutes without proper consideration and testing can have a multiplier effect in terms of all the things that no longer work as intended or expected, and this is multiplied by the number of users times the number of things it breaks for the users. If the QA is thorough, that by itself is not a 5 minute thing, and may take some unpredictable amount of time, due to ensuring that all test cases (this appears to be a highly visual thing that needs to be checked manually, making it that much more of a hassle) pass that are known, and taking reasonable efforts to putting negative tests into place: purposely trying to break things to see if they will break.

Stick to your editorials, and don't go presuming you can instruct developers on how to do their work, or that they should work on your pet project, regardless of how valuable you think it is. At best, this is merely an "enhancement" which is in the eye of the beholder (or in the case of FOSS, in the eye of the beerholder) and may cause more problems than it solves for others. Just for example: how deeply have you analyzed your suggestions for those with color perception issues? There's lots of standard editor themes that for me, even though I have no color perception problems from a medically verifiable standpoint, are horribly intolerable and useless to me, due to sensitivities, but then again, they may work fine for those that are some variation of colorblind. As long as others don't force me to use those themes, ok, I can deal with that. Have you verified that your suggestion is readable to everyone that may use it on the various rendering devices (LCD, plasma, CRT, etc.) and aren't just setup well for a single device? I'm hoping you can see that there may be far more involved than you think when it comes to making a change that affects a lot of people like this.

Exactly, nothing ever takes 5 minutes to fix.

I used to play world of warcraft in my free time, and it was painful reading the forums. There were hundreds if not thousands of clueless kids who presumed to tell the developers that fixing stuff like network lag(!! lol!) was a 5 minute job and that they should get right on it.

I had a good laugh, then a good cry and then I left that place for good.

If I were asked to create a top-10 programmer's words of wisdom list. this would be close to the top. There is simply _nothing_ that takes 5 minutes to fix, no matter how trivial.

It's a running joke where I work. A customer will report a major bug, and someone will comment "How hard can it be? This is surely a 5-minute-fix." Then will all have a good laugh, before trying to figure out what the f*** is wrong this time.

Wait, they do QA???

The screenshot you post isn't even of something that is part of Gnome...

Try posting the "bug" to the right place perhaps? Also, why are you using OSNews to publicize something you think is important, but which no one else really cares about?

The only good thing about SLAB is that you can remove it from the panel.

Why so much venom in these comments?

Technological progress happens because of one thing: INNOVATION. If the linux community puts up such high walls over such a simple feature request, what does that say about it's willingess to adapt and innovate?

So sad.

People get so fixated on the particular technology -- in this case, some aspect of GNOME -- that they have to make the problems fit the technology and not the other way around. How crazy is that?

What I'm trying to say is that instead of asking "How can we make GNOME better?", we should be asking "How can we make DE's (desktop environments) that are more easily modified by the user to fit their custom requirements?" Instead of searching for the elusive, best possible 'one size fits all' result, we need to keep it simple and allow the user more freedom of choice and that includes the choice to more easily create their own solutions.

Linux is free, sure, but the barriers for customization are intimidating to most. When the proposed change can be implemented as easily as writing a Firefox add-on, an xml file, or a cascading stylesheet, maybe then we'll have the creative freedom that is sorely needed.

Edited 2008-07-08 03:58 UTC

Sorry, I missed the underline of the link. My automatic 'error correction' probably filtered out the "or" to make it make sense.

No offense intended.

True, but at the same time, they've already implemented the msn14, which has p2p file transferring; one could always hack around and use that very protocol to deliver video and the client to, rather than save it, stream it directly to some sort of output.

I'm not too sure about the compatibility with existing MSN clients given that it would be hacking up something to do something it was never designed to, as for the quality, I'd say that the video portion of the protocol was designed with QOS in mind so that the quality of the video/audio was as good as it can be.

With that being said, however, Microsoft is apparently going to change their MSN protocol again to what Microsoft's Communication Server uses - IIRC, the Microsoft Communication Server protocols documentation has been opened as part of the 50K page dump recently - so hopefully once Microsoft moves to that protocol for their MSN service, achieving compatibility with it should be alot easier.

Regardless of whether you clearly marked it as "My Take" it was an incredibly ignorant comment to make in the context of the discussion. If you've been complaining for so long about something like that gnome-control-center/gtk+ gripe how have you not even taken the time to educate yourself as to why, exactly, it is something that can't be done within the current toolkit framework? If it's really that big of a deal for you, take advantage of the source code being available for modification and have the change implemented yourself or by someone else on your behalf somehow. If it doesn't have to be a new theme, if it really only requires a clever bit of interface design, then maybe you should educate yourself on how you can fix it to your satisfaction and then give yourself a pat on the back for actually giving back to the community in the true spirit of F/OSS.

I don't want to really even touch the whole V/V jab at the Pidgin developers but seriously, given that you have no idea how monumental a task it is to implement that functionality, why don't you take a step back and realize that you're being incredibly thoughtless. Progress takes time and motivation, and I don't see many people offering to pay any developers to go to work on this particular issue. You might think it's not a big deal, but apparently enough devs think its a big enough deal that they simply have no motivation whatsoever to try to undertake the task of fully compatible voice/video for the various services supported by libpurple.

And please lay off with the lamenting that if you were to bother yourself with fixing something that only you are clamoring about the devs would throw your patch in the trashbin. It is nothing but speculation and unprovable until you actually decide to step up to the plate and test the waters.

I agree with that - problems with Gnome are also problems for all the distro's that use it.

There is a collective "responsibility" to upstream.

You're welcome!
The pleasure was all mine.

When you say "writing your crap everywhere," I assume you're referring to something besides than the grand total of 3 recent posts shown in his comment history?

BTW, thank for proving my earlier post: that OSNews admins completely neglect comment moderation, until someone posts a comment that you personally take exception to.



In other words: you made a post that was 30% relevant content, 70% self-indulgent whining. And now that people have risen to your (blatant) flamebait, you're going to affect victimhood. See that red spot on my shirt? That's my heart bleeding for you.

Almost as amusing as your "I'm taking my ball and going home" blog post (with comments disabled, naturally). I'm frankly amazed that you don't have a writeup on Encyclopedia Dramatica yet - maybe I should remedy that...

Edited 2008-07-06 12:55 UTC

My thoughts precisely. I'm pretty sure that most of the frequent users of this site would be happy with a "Sorry, my bad" but the arrogance on display here is breathtaking. "What? All I did was say fsck this and fsck that and fsck you! I'm right because look at this bug report which is merely a feature request from ages back that obviously is such a high priority in the minds of all! My logic is undeniable!"

Surprised this whole posting and thread hasn't been nuked yet, but at least she's not trying to cover up stirring the hornets nest and getting what she asked for. Yet.

well, I think it boils down to Eugenia simply being at the wrong place at the wrong time. she has stated several times, that she no longer has any interest whatsoever (to put it mildly) in os'es. I don't know exactly why, but she sure seem very bitter about it. now bitter people on OSNews is not exactly a new occurence, sure. but having a moderator show not only a total disinterest, but also what comes across as resentment towards the actual subject matter on which OSNews bases it's existance is simply a bad recipe.

I can't help but think that for both OSNews sake and the sake of Eugenia herself, she should stick to her gadgets and videography from here on.

I just read her blog post and I was shocked and offended by what I read. Eugenia needs to resign from OSNews. Her credibility is in question and her bias obvious. Get someone else.

Excuse me, sir, but please don't generalize your comments to apply to all OSNews editors. You don't know what really goes on in the database and with your fellow users, but I do. Since implementing peer moderation, we've only done about .3% of ALL moderation on the site. That's POINT three percent.

It's hardly fair for you to take your frustrations with one person out on everyone.

You don't moderate, yet my comments went from +8 +5 and +2 to all being -5.

Yep, you don't moderate.

Seems like you still have a problem even if you yourself won't admit it.

Thanks.

I am a developer too and because of that, I would not compare my work to the KDE/Gnome developer's work! KDE/Gnome developers deserve a lot of respect... I did not give my work to the community as they did. Maybe a little humbleness is missing, isn't it?

What's wrong with you Eugenia? The last two posts in OSNews are full of your modded down comments!

How can you ban or "temporarily disable" someone's account because he is posting "crap everywhere"...

Will I have the same possibility of ban or disable the accounts of people that disagree with me? that mod me down? that reply me with things I do not want to hear? I do not think so... Thus, someone here is abusing the power she has in OSNews.

Edited 2008-07-06 15:24 UTC

Good, and do all of us a favor and don't ever come back. I just read your blog, and hope your arse never comes back here again.

Agreed. Though many applications (e.g. firefox, eclipse) support resuming where they were, so nowadays I don't mind shutting down and restarting as much as before.



Multi-monitor support has improved greatly with 8.04; I was about to switch back to Windows just for this reason after years of using Linux as my main OS, but luckily 8.04 solved most of my issues so I didn't have to. There is a nice UI available too. YMMV for different models, but it is definitely an area that has been focused on, which also means that if you still encounter problems, you have good chances of getting them fixed by filing bug reports.

My biggest issues have been with multilingual input (SCIM) and audio with proprietary software (flash, skype).

Ah, very interesting the aMSN support for video, along with limited voice support, I hadn't heard about all that. I take it the support is cross-platform? Worked just fine setting things up over here on my Windows install but I don't have anyone handy at just this moment to test the functionality first-hand. aMSN is licensed under the GPL apparently so code-sharing should be quite possible, I think the Pidgin devs will likely lift it once they start trying to implement full v/v support. Kinda got me wondering if the aMSN team got the code from the gaim-vv projects efforts?

I doubt it gaim-vv has been dead for a long time now.

I've known about aMSN for awhile now.

But I just found out about GYachE Improved today.

GYachE Improved just blows Pidgin's Yahoo support out of the water. It can literally be compared to Y!mLite for *nix.

Pidgin is about to get kicked off my desktop.

Edited 2008-07-06 01:39 UTC

GYachE Improved really sucks because of the GUI. It's simply a nightmare. There are so many wrong things that I don't know where to start. I'd rather use meebo than this.

It seems to have a lot of Yahoo features that lack from Pidgin like buzz, fade colors, login as invisible (implemented properly), some audible and IMviroment support, webcam support and even something that puzzles me ... Yahoo chat support, but these features seem to be put one on top of the other to a degree that you struggle to use this application.

Edited 2008-07-06 07:21 UTC

I'm not offended but he's just using excessive profanity to mask the fact that he has nothing substantial to say.
The blog being incredibly trite and boring is just a side-effect.

That is hilarious! I can't tell if "Ubuntu Christian Edition" is a serious distro, or simply an Onion joke that went wwaaaaay too far.

I think it lacks malware to begin with.

Linux distributions ship mostly open source code ... some of them go to quite some lengths to ship only open source code.

Open source code means that anyone can compile the code for themselves, to ensure that the source they see creates the package as shipped.

Open source code also means that any developer/programmer who has an interest in actually using that can look at the source and assure themselves that there is no malware within it.

So, as an end user of a Linux distribution ... if you adopt a policy of installing only open source code from the distribution's repository ... you then know that you are using code that developers/programmers have the ability to audit and they are using it themselves. No-one is going to use code where anyone can see how it works and yet it still contains malware.

Hence ... that code doesn't have malware in it. As asgard said it far more succinctly: "I think it lacks malware to begin with."

In other words ... malware is necessarily constrained to closed-source programs only for its delivery.

Edited 2008-07-06 10:28 UTC

True enough.

However, it's a bit tricky to write a piece of software to detect something that doesn't exist. Yet.

Essentially, all Windows anti-malware applications work on two principles. First, a database of known malware and some mechanism of detecting it, and secondly a database of known hooks said malware commonly uses. Most, in fact, only have the first. Complex ones for advanced users who know what they're doing have the second.

Since there isn't any malware yet, and there's really no way to know exactly where this hypothetical malware would install itself, it's kind of impossible to write an anti-malware tool.

That said, there are rootkit scanners, intrusion detection tools, and tools to check the integrity of installed software. They're all targeted at servers, of course, since that's where the only existing Linux malware is.

There's rkhunter and chkrootkit, which it doesn't hurt to run once in a while.

But seriously, there's virtually no need for malware detection; people just aren't targetting <1% of the desktop market with that sort of thing.

He did offer a solution, for example for LSB. That solution was to standardize (at least between the major distros*, because no one** cares about Gobolinux) the binaries. He was also supportive of the synchronized releases system.

Both of which won't happen because of "features" and "product differentiation" (http://en.wikipedia.org/wiki/Product_differentiation). For example, how many new features can libc get in 2 years' time? (and why would you want new features in libc?!)
Some things should be standardized, and you could only add on top of that.

Then, on the next cycle, you could change the versions.

Not glamorous, but ultra effective. And remember kids, different base libraries version don't offer more choices, they hinder them. What you want is more higher quality application alternatives (for the "free software is about freedom to choose" fanatics ).

*Red Hat (aka Fedora, CentOS + other clones), SUSE (aka openSUSE), Debian (+ derivates), Ubuntu (which has marketing clout), Linspire/Xandros. Gentoo can't be counted in because of the development policy, Slackware has 3 people developing in, PCLinuxOS also has few people developing it. Basically if Red Hat, SUSE, Debian and Ubuntu will agree to something, everyone would follow in due time.
**No one = 1% or less of the Linux desktop population.

Edited 2008-07-07 07:33 UTC

If the private directory is mounted the encrytption is not in use.

If the private/revoke key is on the same machine it's similar not very usefull.

A default simple "chmod -R go-rwx /home/*" which even root can't alter only the owner is the same security through obscurity imho.

There simply are way more attack tools for *nix then security products :-)

Windows is the only OS that i know off that doesn't give you during the instalation ore soon thereafter, the opportunity of installing a whatever flavour free fullblown development environment.

Edited 2008-07-06 15:16 UTC

Without encrypted swap and temp, it's pointless to encrypt files and filenames.

It's as easy as steal -> put drive in new device -> strings /dev/swap partition.

The OS is even irrelevant.

The only good answer (currently) is a drive that self encrypts, OR a fully encrypted disk.

Unless you just want to obscure files, and don't care if the contents are retrievable on theft. In that case, encrypting a file system without a signature is fine as long as you don't keep shell history or event logs anyway.

:-D

I'm eager to play with Truecrypt whole disk encryption, but LUKS for everything but /boot seems to work very well.

Replied to wrong thread..

Edited 2008-07-06 15:32 UTC

FYI Totem BEEN having gstreamer and Xine backends for years.

Using both gstreamer and Xine backends on the same frontend without removing one of them is fairly recent.

Well Said.
You got my point exactly.

As I see it, the latest release of Ubuntu is pretty poor. There have been numerous cases cited where things that worked in V7.x now fail miserably.
This is akin to the pile of rotting seaweed that Fedora Core 4 was.
We should all face up to the facts that once in a while those who release the distros we know and love sometimes throw us a curve ball.
This release might very well be that for Ubuntu.
I'm sure they will get it right for the next one (after Ibex possibly) as the feature list is somewhat bare. Perhaps they should not implement any new features just concentrate on fixing the bugs with the current one.

I also hope that some of the Ubuntu 'fan-boys' realise that Ubuntu & 'apt-get' is not the answer to 'Life, the Universe and Everything' ( I read this somewhere recently but can't remember where and found it very 'apt'...pun intended)
Edited to fix two typos

Edited 2008-07-06 16:43 UTC

True, i rather prefer OpenSuSE 11.

Eugenia filed that bug years ago. In all this time she could have submitted a patch for it, but didn't.

Do you really think that the people capable of developing a theme are also capable of working on the "dozens of errors that could be fixed or features that could be implemented"?

They aren't. Completely different skillsets. What one group is doing doesn't prevent the other group from doing something else.

The point of encrypted partitions is that if, say, you get your laptop stolen, at least the confidential information in it can't be easily retrieved. This is not true for mere permission-protected files. Having an encrypted directory instead of an encrypted partition is, of course, much more flexible and convenient. If anything, one may wonder why it was not available before.

2:
Encryption is very useful when person trying to read your data has physical access to the machine. You can log in as a root in any kind of environemnt (like LiveCD, or just another machine) and permissions won't help you a lot.