Let's put this straight: This is about people running software that they're not entitled to run. Simple as that. Yes, they made a choice to do so. First, everybody was whining when Microsoft tried to more or less disable those systems, so MSFT stopped the practice. Now you're whining because MSFT is not handing out free software to those leeches? Get a fucking grip on reality! In my opinion, everybody complaining about things like those is more part of the problem than anything.
Things I am still using for more or far less obvious reasons:
KDE: With version 4, KDE went the all-too-common way of going for bling over usability. SSL certificate management was still broken last time I checked. Konqueror fails at a lot of things (MSDN among others), which could be vastly improved by switching to WebKit (an IRC client I'm using uses it for previews, and those look better than what Konqueror makes of them!), but that would require pissing off the KHTML team. The print setup dialog is suffering from amnesia (ever tried duplex printing?) since the old but working KPrint subsystem was unmaintained.
Firefox: Pretty much the same as KDE. Useless bling and features are getting hacked into a huge legacy codebase without any regard for regressions, privacy considerations, or security aspects.
xine: Nice program, pretty good architecture, but for all it's worth, it's unmaintained. No seeking in FLAC or WavPack, Matroska support is sketchy (and that's stretching the definition), and overall it would need a few dozen people with half a clue to go through the myriad demultiplexers and decoders and just throw out most of it. Along with xine, I will just put every KDE multimedia application on the list, since those need some backend, and all Phonon backends but xine seem broken beyond recognition.
OpenOffice.org: Oh. My. God.
KOffice: Oh. My. Other God.
XFree/X.org: More of a systemic disease. Hopefully Google will realize that it sucks ass when they're making their OS and create something that's not quite as ugly.
This list is by no means comprehensive.
Bug reports are answered in several ways:
Not al all. That's the most prevalent way of handling things. Maybe after a few dozen "me too"s, someone will pretend to actually read the BTS.
"We do not fix UI bugs." With regards to Mozilla.
"Patch it yourself." This is my absolute favourite. How about "Uhhh... no"? Chances are that I haven't written a line of (obscure language of choice) in (huge number) years. Also, I am not interested in trying to apply the patch every time the distribution or the upstream fudge with their end of the package.
There are a few exceptions, but that's my general experience.
(This also is a reply to the much more obvious troll above.)
This is not about "making firefox less compatible", this is about vulnerabilities. Font renderers like e.g. FreeType2 had the odd glitch here and there, and there is no guarantee that there won't be others. Hey, if bugs and vulnerabilities can't emerge from formerly (reasonably) safe code like, say, JavaScript support in Firefox (you can stop laughing... now), then it would only be fair to assume that those libraries could be exploited eventually.
The renderer currently takes any given downloadable font without asking, and stuffs it down another library's throat without asking. I would at least like to be asked if I agree to that. The alternative would be obnoxious "transparent" proxies or firewalls, or a virus scanner, but why depend on stuff like that when a simple per-(action|page|session) browser setting would suffice?
It's like Type3 fonts in Postscript. Just make a custom font with the glyphs permutated a bit, transform the text accordingly, and hey presto, copy is worthless. Or how about having complete paaragraphs or pages in a single glyph?
I have to agree there. Marvell chips have given me nothing but grief, both in Windows and Linux, both network and ATA. They made me buy a SATA DVD writer to replace the fully functional PATA device, as well as a Silicon Image-based SATA controller to attach harddisks (couldn't use the Intel controller, because MSI still insists on placing the "good" SATA connectors under the GPU cooler).
All in all, I'm happy about anything bad happening to Marvell.
As a broad generality:
The onboard and inexpensive SATA and IDE controllers seem to have problems detecting a failure or even which drive is failing. They also don't seem to have a good way to report to the OS that a drive is having troubles. It is also common for them to not have a way to rebuild the mirror from a single drive. It is mind boggling when the build functions wipe both drives!!!
... which is why I wouldn't even consider using the RAID functionality of those controllers. I'd rather use pure software RAID like e.g. the Linux kernel offers, and which I've been using for several years without a single hitch. Simple RAID1 controllers aren't expensive, especially when compared to the peace of mind and the value of the data they're protecting. With the competent products, you will get things like monitoring software, really audible alarms (we're talking out of the rack, through two walls and a wall-filling AC unit here), matching hotswap enclosures (most onboard fakeraid controllers even fail at that), "warm" spares that are only even spun up when a live drive fails, schedulable maintenance tasks that run independent of the OS, etc.
If someone skimps out of buying the right hardware for the job, their data can't be worth all that much, right?;-)
And testing if the notifications work should really, really be part of the setup process.
The bad part of recovery is when you are working with a drive that is seriously in failure mode and getting worse you are in a race against time getting the data off the drive.
Right, but that is where the whole RAID comes into play: When a drive goes into failure mode, you don't give a damn (and switch to hot/cold spare). You also have off-medium backups, right?
The whole problem has at least two tiers: first comes a reliable solution you work with on an everyday basis, which is a RAID1. Let's just assume that the OS is halfway competent at not randomly killing stuff there, and maybe add a battery backup on the controller to deal with the odd power failure.
However, you're also supposed to have a sufficiently recent backup on another medium that the OS or a power surge is not free to screw up on a whim, even if it's just a pair of external HDDs that are alternated and kept disconnected from the system (remember kids: harddisks are really dirt cheap!). It doesn't even really matter what kind of FS you're using on those, since the chances of both the RAID and the backup medium failing in quick succession should be sufficiently small.
In my experience, NTFS and other journaling file systems mostly break in rather catastrophic conditions, i.e. power loss or kernel-level system crash. FAT on the other hand tends to corrupt silently out of the blue, and wind up with lots of inconsistencies after a few months of seemingly normal use.
As an added bonus, you will get severely degraded performance when dealing with "a lot" of files, e.g. a Lightroom archive with around 15,000 pictures in it. I use rsync to back those files up (to a small fileserver with a 2-channel 3ware controller by the way). On a FAT partition, rsync used upwards of half an hour just to determine which files to back up, while the same operation is virtually instantaneous on an NTFS partition. Similar things occur for operations like file system checks and defragmentation.
Just a thought. Make simple games that are fun to play? Use a system for what it's capable of, and not for what it isn't.
Hell, I'd suppose Wing Commander 1 or 2 could be ported to the Wii (with some control changes, maybe a communication/navigation system à la i-War to get around the lack of a keyboard, and that'd not even be a low-profile game.
Regardless, if money is there worry, they should release an expansion pack, Not a sequel.
Their customers would then be whining about how the expansion pack is too expensive, has too few maps, the quality is generally lacking, and waaaah waaaah waaaah. It is really easier and more profitable to market a proper sequel. Gamers are an especially whiny bunch that gets upset and all pee-in-the-eyes at the smallest issue, and no amount of goodwill from the game makers will fix that.
Version numbers are hardly a reliable indicator of quality when all/most of the code has changed behind the scenes.
Instead of all the vitriol, route around the problem. Use the Gnome/desktop agnostic counterparts of the programs until the devs fix their packages.
Still, the desktop environments are marketed as new pinnacles in shinyness and productivity, and breaking functionality that was very much alright in the predecessor (which in turn had a bunch of pending bugs that were supposedly going to be fixed in the next big release two years down the road) is indeed a regression. If this was some dumb little hobby project that lets cute kitties run across the screen, I'd be alright with that. If it is an environment that claims to be fit for production use, or has some shiny new major version number with some near-invisible blog entry begging users not to use it, that's idiocy.
Re:Will we get another "don't use me yet" "release
on
KDevelop4 Beta 3 Released
·
· Score: 0, Offtopic
It's a new app, there is no such thing as a regression. Same applies to many other KDE4 progs.
"many other KDE4 progs" are meant to be improvements of the originals, it says so right in the version number. I can't see any valid reasons to accept the state a lot of those major releases are in.
KDE 4 itself has rather ugly corners, like e.g. SSL key management, printer management (kdeprint was apparently unmaintained, so the developers decided to stick with whatever the Qt team belched up).
Amarok 2 had a major release that consistently failed to build a proper library (randomly missing tracks depending on the phase of the moon during import, ugly problems with extracting metadata), was prone to crash, and relies on Phonon for which there doesn't seem to be a reliable backend (xine can't even seek in FLAC of WavPack, gstreamer integration is sketchy at best).
... which is pretty much it for multimedia.
Calling Plasma a rough takeoff would be too kind. But hey, it can haz flashy effects!!!1!
The last release candidates of KOffice 2.0 had ugly font rendering, an extremely... esoteric font/style management, and the mentioned missing settings dialog. So tough luck if you want to e.g. show formatting characters like linebreaks.
There are probably more which I chose to forget about just to keep the blood pressure low.
Will we get another "don't use me yet" "release"?
on
KDevelop4 Beta 3 Released
·
· Score: 0, Flamebait
As seen with KDE 4.0 ("Can't even save printer settings"), and now KOffice 2.0 ("Who needs a settings dialog anyway?")?
Seriously, someone needs to find who wrote the fucking memo that says you can have all the regressions you want as long as your applications use Qt4.
This is ESPECIALLY true when dealing with CAT7 or STP. On a 20Mb line (Probably a 100Mb link) the chances of having a problem though are pretty low provided you terminate it cleanly.
This is assuming the OP is actually dealing with an ethernet connection entering the premises. It's far more likely he's dealing with some kind of serial link that is properly terminated, and he just needs to link it with the rest of the network (i.e. hook it up to a router). In that case, there are no termination issues or anything. Hell, I guess a length of Cat5e with cheap connectors would do the trick.
There are some vandalism-proof keyboards with stainless steel surfaces, which should survive a lot of "soft" sterilization methods like surface disinfectants, highly UV resistant, basically anything less demanding than an autoclave. An example can be found here: http://www.industrialkeyboard.com/html/vandalism-proof_stainless_stee.html (even comes with an integrated trackball).
That is why this is a good move. Face it, OOo is the horrible bastard child of "development by committee" that has not really moved forward for several years now. The text processor part is almost usable, but the rest is just a bloated bugfest. Having this as the primary association to the document format cannot be a good thing.
In the long run, I hope alternative tools will emerge (no, KWord does not count yet, it still produces rather interesting results on most documents that are not walls of text) that are not based on ancient mystical code someone found in a cave.
Slashdot Mirror
Let's put this straight: This is about people running software that they're not entitled to run. Simple as that. Yes, they made a choice to do so. First, everybody was whining when Microsoft tried to more or less disable those systems, so MSFT stopped the practice. Now you're whining because MSFT is not handing out free software to those leeches? Get a fucking grip on reality! In my opinion, everybody complaining about things like those is more part of the problem than anything.
If he spent 331 workdays browsing porn, he seriously needs to renegotiate his position, because he's getting screwed over holiday-wise.
Things I am still using for more or far less obvious reasons:
This list is by no means comprehensive.
Bug reports are answered in several ways:
There are a few exceptions, but that's my general experience.
Fundies like these Taliban idiots, never stick around for the discussion.
Best. Comment. Ever.
(This also is a reply to the much more obvious troll above.)
This is not about "making firefox less compatible", this is about vulnerabilities. Font renderers like e.g. FreeType2 had the odd glitch here and there, and there is no guarantee that there won't be others. Hey, if bugs and vulnerabilities can't emerge from formerly (reasonably) safe code like, say, JavaScript support in Firefox (you can stop laughing... now), then it would only be fair to assume that those libraries could be exploited eventually.
The renderer currently takes any given downloadable font without asking, and stuffs it down another library's throat without asking. I would at least like to be asked if I agree to that. The alternative would be obnoxious "transparent" proxies or firewalls, or a virus scanner, but why depend on stuff like that when a simple per-(action|page|session) browser setting would suffice?
It's not a joke. However, since IE doesn't seem to support it, we're unlikely to see custom fonts being abused that way.
It's like Type3 fonts in Postscript. Just make a custom font with the glyphs permutated a bit, transform the text accordingly, and hey presto, copy is worthless. Or how about having complete paaragraphs or pages in a single glyph?
In about:config, set gfx.downloadable_fonts.enabled to false and restart the browser.
I have to agree there. Marvell chips have given me nothing but grief, both in Windows and Linux, both network and ATA. They made me buy a SATA DVD writer to replace the fully functional PATA device, as well as a Silicon Image-based SATA controller to attach harddisks (couldn't use the Intel controller, because MSI still insists on placing the "good" SATA connectors under the GPU cooler).
All in all, I'm happy about anything bad happening to Marvell.
As a broad generality: The onboard and inexpensive SATA and IDE controllers seem to have problems detecting a failure or even which drive is failing. They also don't seem to have a good way to report to the OS that a drive is having troubles. It is also common for them to not have a way to rebuild the mirror from a single drive. It is mind boggling when the build functions wipe both drives!!!
... which is why I wouldn't even consider using the RAID functionality of those controllers. I'd rather use pure software RAID like e.g. the Linux kernel offers, and which I've been using for several years without a single hitch. Simple RAID1 controllers aren't expensive, especially when compared to the peace of mind and the value of the data they're protecting. With the competent products, you will get things like monitoring software, really audible alarms (we're talking out of the rack, through two walls and a wall-filling AC unit here), matching hotswap enclosures (most onboard fakeraid controllers even fail at that), "warm" spares that are only even spun up when a live drive fails, schedulable maintenance tasks that run independent of the OS, etc.
If someone skimps out of buying the right hardware for the job, their data can't be worth all that much, right? ;-)
And testing if the notifications work should really, really be part of the setup process.
The bad part of recovery is when you are working with a drive that is seriously in failure mode and getting worse you are in a race against time getting the data off the drive.
Right, but that is where the whole RAID comes into play: When a drive goes into failure mode, you don't give a damn (and switch to hot/cold spare). You also have off-medium backups, right?
The whole problem has at least two tiers: first comes a reliable solution you work with on an everyday basis, which is a RAID1. Let's just assume that the OS is halfway competent at not randomly killing stuff there, and maybe add a battery backup on the controller to deal with the odd power failure.
However, you're also supposed to have a sufficiently recent backup on another medium that the OS or a power surge is not free to screw up on a whim, even if it's just a pair of external HDDs that are alternated and kept disconnected from the system (remember kids: harddisks are really dirt cheap!). It doesn't even really matter what kind of FS you're using on those, since the chances of both the RAID and the backup medium failing in quick succession should be sufficiently small.
In my experience, NTFS and other journaling file systems mostly break in rather catastrophic conditions, i.e. power loss or kernel-level system crash. FAT on the other hand tends to corrupt silently out of the blue, and wind up with lots of inconsistencies after a few months of seemingly normal use.
As an added bonus, you will get severely degraded performance when dealing with "a lot" of files, e.g. a Lightroom archive with around 15,000 pictures in it. I use rsync to back those files up (to a small fileserver with a 2-channel 3ware controller by the way). On a FAT partition, rsync used upwards of half an hour just to determine which files to back up, while the same operation is virtually instantaneous on an NTFS partition. Similar things occur for operations like file system checks and defragmentation.
... game experience?
Just a thought. Make simple games that are fun to play? Use a system for what it's capable of, and not for what it isn't.
Hell, I'd suppose Wing Commander 1 or 2 could be ported to the Wii (with some control changes, maybe a communication/navigation system à la i-War to get around the lack of a keyboard, and that'd not even be a low-profile game.
Regardless, if money is there worry, they should release an expansion pack, Not a sequel.
Their customers would then be whining about how the expansion pack is too expensive, has too few maps, the quality is generally lacking, and waaaah waaaah waaaah. It is really easier and more profitable to market a proper sequel. Gamers are an especially whiny bunch that gets upset and all pee-in-the-eyes at the smallest issue, and no amount of goodwill from the game makers will fix that.
Version numbers are hardly a reliable indicator of quality when all/most of the code has changed behind the scenes.
Instead of all the vitriol, route around the problem. Use the Gnome/desktop agnostic counterparts of the programs until the devs fix their packages.
Still, the desktop environments are marketed as new pinnacles in shinyness and productivity, and breaking functionality that was very much alright in the predecessor (which in turn had a bunch of pending bugs that were supposedly going to be fixed in the next big release two years down the road) is indeed a regression. If this was some dumb little hobby project that lets cute kitties run across the screen, I'd be alright with that. If it is an environment that claims to be fit for production use, or has some shiny new major version number with some near-invisible blog entry begging users not to use it, that's idiocy.
It's a new app, there is no such thing as a regression. Same applies to many other KDE4 progs.
"many other KDE4 progs" are meant to be improvements of the originals, it says so right in the version number. I can't see any valid reasons to accept the state a lot of those major releases are in.
There are probably more which I chose to forget about just to keep the blood pressure low.
As seen with KDE 4.0 ("Can't even save printer settings"), and now KOffice 2.0 ("Who needs a settings dialog anyway?")?
Seriously, someone needs to find who wrote the fucking memo that says you can have all the regressions you want as long as your applications use Qt4.
And drop some legacy systems (X comes to mind) along the way.
This is ESPECIALLY true when dealing with CAT7 or STP. On a 20Mb line (Probably a 100Mb link) the chances of having a problem though are pretty low provided you terminate it cleanly.
This is assuming the OP is actually dealing with an ethernet connection entering the premises. It's far more likely he's dealing with some kind of serial link that is properly terminated, and he just needs to link it with the rest of the network (i.e. hook it up to a router). In that case, there are no termination issues or anything. Hell, I guess a length of Cat5e with cheap connectors would do the trick.
Cut Russia and China off the internet for a week and see what it does.
I'm a bit confused, don't the Republicans have Dumbo as their mascot?
To add the two, I got an inline ad to this in this thread. (Konqueror helpfully offered to open it in KWrite.)
'nuff said.
There are some vandalism-proof keyboards with stainless steel surfaces, which should survive a lot of "soft" sterilization methods like surface disinfectants, highly UV resistant, basically anything less demanding than an autoclave. An example can be found here: http://www.industrialkeyboard.com/html/vandalism-proof_stainless_stee.html (even comes with an integrated trackball).
That is why this is a good move. Face it, OOo is the horrible bastard child of "development by committee" that has not really moved forward for several years now. The text processor part is almost usable, but the rest is just a bloated bugfest. Having this as the primary association to the document format cannot be a good thing.
In the long run, I hope alternative tools will emerge (no, KWord does not count yet, it still produces rather interesting results on most documents that are not walls of text) that are not based on ancient mystical code someone found in a cave.