Alan Cox Quits As Linux TTY Maintainer — "I've Had Enough" 909
The Slashdolt writes "After a stern criticism from Linus, the long-time kernel hacker Alan Cox has decided to walk away as the maintainer of the TTY subsystem of the Linux Kernel, stating '...I've had enough. If you think that problem is easy to fix you fix it. Have fun. I've zapped the tty merge queue so anyone with patches for the tty layer
can send them to the new maintainer.'" A response to a subsequent post on the list makes it quite clear that he is serious.
Linus (Score:5, Interesting)
Linus is brilliant. He is funny. Most days I really agree with anything he has to say.
However, he has butted heads with people in the past. Perhaps this is just human nature and unavoidable from time to time. Linus isn't perfect, nor always right. I thought he was really unfair to Con Kolivas when he drove Con away.
Interesting... (Score:4, Interesting)
I'm curious about how projects, in general, fare after someone with rather intimate knowledge leaves for whatever reason. I'm not being specific to Linux; you gotta think some of the kernel developers of Windows have left over the years. That's gotta be hard on the next person regardless of project; "here's his code, all three million lines of it. Oh, he seemed to like Pascal syntax so he wrote all these macros to make his C++ code look like Pascal. Good luck!"
On slashdotting... (Score:5, Interesting)
WHY can't lkml.org's mailing list retriever handle a slashdotting?
Its not like the flashcrowds are all THAT big.
Re:Interesting... (Score:5, Interesting)
Indeed, ive jumped into quite a few projects replacing a previous programmer. Some were experienced and reading their code was really interesting, others were fired for being incompetent and I ended up rewriting most of their code.
In any cases, the first few days, weeks or months depending on the size of the code are spent studying the structure rather than actually coding.
Re:Interesting... (Score:5, Interesting)
ive jumped into quite a few projects replacing a previous programmer. Some were experienced and reading their code was really interesting, others were fired for being incompetent and I ended up rewriting most of their code.
In any cases, the first few days, weeks or months depending on the size of the code are spent studying the structure rather than actually coding.
The differences here are that A) this is an open-source project, B) this is a *HIGH PROFILE* open source project, and C) Alan was the maintainer, not sole coder (so he both coded, and accepted patches from others.)
It's possible that Alan was the only one who knew anything about the TTY code and how it worked, but I'd doubt it. I'd be really surprised if the new maintainer comes into the role cold.
Re:Not diplomatic (Score:1, Interesting)
Could anyone shed some light... (Score:5, Interesting)
Re:Linus (Score:5, Interesting)
Serious question: what percentage of the current Linux kernel was actually written by Linus?
Re:Linus (Score:3, Interesting)
BSD has terrible driver support compared to Linux.
If you don't like the way Linus maintains the Linux kernel, then fork it and maintain your own branch. That's a lot easier than messing around with a different kernel that doesn't have half the driver support that Linux has. All you have to do is take what's there in Linux, then change the things you don't agree with (like dumping ALSA and moving back to the new OSS).
It'd really be interesting if someone seriously made a Linux fork to compete with Linus.
Re:Linus (Score:5, Interesting)
I'm reading the thread. This is my take.
Cox -> submits code which apparently caused a bug
User -> Reports breakage
Cox -> Can't replicate breakage and asks user for debug info so he can fix it.
User -> Says they don't know what to debug for, but is willing to work with Cox.
Linus -> Jumps in and calls Cox's code a buggy piece of shit before any debugging took place, and before it is established if the code is buggy or not.
Cox -> Continues to troubleshoot the issue.
Linus -> Flames Cox personally and says Cox is unwilling to work on the issue.
Cox -> Takes his ball and goes home, except in this case, it is OSS so he doesn't really take any ball with him. He just leaves.
Re:Linus (Score:5, Interesting)
Harry Truman used to say he didn't want to talk to experts because an expert was just someone who didn't want to learn anything more on a subject because if he did, then he'd find out he didn't know it all and wasn't an expert any longer.
Re:Linus (Score:5, Interesting)
The quality of a programmer is often proportional to his ego.
Be careful: Humans confuse cockiness with expertise [newscientist.com].
Re:Thanks (Score:5, Interesting)
How about "Nice work Linus, you'll have the entire kernel back to yourself any day now, I'm sure"
Back three years ago I was sure I'd never leave. Now, I was no kernel dev, but I found out what it was like to try. In the meantime I grew up, and realised there's two sides to Linux.
Go and have a look at forums.gentoo.org, where you'll see both at work. I gave up too. For a long time I thought, through contributions and advocacy, I'd help Linux make some real headway in the Server and desktop market. Eventually I came to believe that it would never be big, it'd just mean more communities and more infighting and little real progress.
So I'm sorry, Alan. I'm really sorry, but you've made the right move. Thanks for everything.
Re:Linus (Score:5, Interesting)
I read the LKML for years.
Ingo did write the new scheduler, at the request of Linus. Ingo didn't make personal attacks on Con.
Linus was the one for years who said Con was wrong about scheduler theory. Ingo admitted Con was correct, but Linus wouldn't admit he was wrong. Linus asked Ingo to write a new scheduler, basically ignoring the one Con had submitted.
When several people pushed to include Con's scheduler (which at that point was called Staircase) Linus made more personal attacks and wrapped it up saying that Con couldn't be trusted to support his work.
Ingo had nothing to do with that.
Re:Should I? (Score:3, Interesting)
I once had a bear do exactly that! We were camping at Merced Lake in Yosemite, and had six-man-weeks of food for three of us to take a two week backpacking trip. We put it 20' up a tree, on the end of a 10' limb. In the middle of the night, we heard loud stomping sounds, a bunch of scrapping sounds, then a rope-through-pulley sound, and a loud crash. A damned black bear had jumped from the tree, caught our food bag on the way down, using our rope to slow his crash, and squashed a small tree when he landed.
My friend had a great idea. "Let's go get our food back." I don't know how I let him talk me and my other buddy into it, or why I was the one in front with a dying flash light, as we followed the munching sounds into the woods. Suddenly, the bear stood up on both back legs with his arms out, and I don't know if the roar I heard was the bear, or just the blood rushing to my head. I turned around to ask if we should run, but there was no one there! I caught up to my brave friends at the lake, where we discussed wading into the freezing cold lake vs. making a stand there.
I must say, it didn't occur to me to check out the bear's penis, but he was waving everything else at me!
The butt hurt started with KDESU? (Score:4, Interesting)
Seriously? KDESU is broken, in the first place.
https://bugs.kde.org/buglist.cgi?quicksearch=kdesu
Re:Linus (Score:3, Interesting)
The OP wasn't specific about which BSD variant he would switch to; neither is de Raadt representative of how all the BSDs are managed. FreeBSD's governance structure has worked marvelously well over the years, so I would say it's unfair to implicitly lump every BSD under the same umbrella as OpenBSD and its management.
(I realize that may not have been your intent, but you used the same generalizations as the OP. Just being fair here. :)
Drag'n'drop (Score:4, Interesting)
Drag and drop is quite convenient. It is also a security chasm. Once you can drag and drop one thing then you want to be able to drag and drop anything. In the case of arbitrary file formats not only must you implement code to check the incoming data stream (thus exposing yourself to all of the security considerations of "how many different ways can someone try to wax my process of checking the incoming data stream?") but you must consider that a data stream which is valid using one codec algorithm may cause a fault using another codec algorithm. Competing algorithms exist for many data format structures and the presumed same data format may have three or four codecs at use between X, the WM, a monolithic app like a web browser, and a devoted data editor (eg. GIMP), and even a devoted data viewer (eg. a multiformat display application). It isn't the simplest consideration.
With so much of the problem and criticism with the reigning proprietary OS being security related the open source community has tried to remain a little more focused on security related issues. Combine that with the difference in conceptual organization--F/OSS guys don't get paid to go to in house meetings together--and it is completely logical that something as "simple" as drag and drop is not implemented across largely unrelated application development groups.
Within a particular desktop environment using apps which were written specifically for that desktop environment (often referred to as a desktop suite) there is probably a more consistent end user experience.
It is the culmination of (years of) similar situations which has brought many rifts in major F/OSS development groups.
I find myself personally familiar with the situation which caused Alan to leave. The difference is that Alan has enough financial backing and social connections behind him that he likely will not end up living on the streets.
Can you imagine a headline,"Major developer sick and tired of political crap, leaves development group, will take up a section of cardboard on the sidewalk just down the block from Slashdot's HomelessinLaJolla"?
Why is the TTY subsystem still in the kernel? (Score:3, Interesting)
I'm surprised that the TTY subsystem is still in the kernel. By now, it ought to be in user space. It's rare today that a serial port is attached to an actual terminal (let alone a real Teletype [sourceforge.net]), and separating the serial port driver from all the backspacing and line handling stuff would make both parts simpler. Most of the time, the TTY stuff in the kernel just gets in the way of other uses of serial ports.
They've been separated in QNX for a decade, for example.
Re:Linus (Score:5, Interesting)
That is actually just an UI bug in the door. If you want people to push a door, you should use a handle that is like a plate, where you can easily put your hand against and push it. If you want people to pull the door open, you need to use vertical rod as a handle, where people can easily grab on to pull it. With this very small change, you don't even need to push/pull texts on the doors.
Also, doors should be always pulled when you go in and pushed when you go out. That makes exiting the building easier in case of emergency (people don't rush to the door and jam it, preventing anyone from pulling it open.) and also when people are trying to get in and out at the same time, the person outside is more capable of keeping the door open for the person going out (it is better that people first get out, before new people get in, because inside there is a limited space, while outside contains usually a lot more room). Also outside usually contains more room for pulling, while the inside often has a wall that limits the space for pulling, especially if you want to keep the door open for someone else.
Re:Theo (Score:5, Interesting)
Oh boy, I can't agree more with this statement. I realize this is anecdotal, but here's my own personal experience...
In the Gentoo Freenode channel, it was difficult to so much as type without bumping into someone's ego. Participants were generally rude, crude, and--perhaps most ironically--threatened to kick people for using swear words. I have no idea if this has changed at all in the 4 or 5 years since I last joined... but it wasn't a pleasant experience.
Contrasted with the various FreeBSD channels I used to join infrequently, the experience was on the whole much better. People were friendlier, had a sense of humor, were helpful, and didn't get their underpants in a dozen knots over something incidental like a single, mostly unoffensive swear word. Again, it's been years since I participated in that as well and perhaps the FreeBSD channels have changed...
Personally, I doubt it. It's a cultural difference, I think. The BSD crowds seem more product-driven (let's get Y done) versus some Linux distros that seem process-driven (I don't like how you're doing X and it doesn't matter if we're making Y).
This, of course, is purely anecdotal. You don't have to agree with it because it was my personal experience, and as such, FreeBSD folks have come off to me as MUCH more friendly and cooperative.
Re:Thanks (Score:3, Interesting)
Agreed. Alan, thank you for all the hard work for so many years. You deserve a vacation.
Re:So long and thanks for all the code. (Score:5, Interesting)
Actually, and bizarrely enough, there is a high level FreeBSD developer named Alan Cox too
http://en.wikipedia.org/wiki/Alan_Cox [wikipedia.org]
Re:Should I? (Score:4, Interesting)
Once we camped out near a river, in known bear territory somewhere in central California. We hadn't seen any bear tracks, but put our food up in a nearby tree anyway (because that's just what you do in bear territory).
At around 2 AM that night, we awoke to hear the sound of large animals moving in our campsite, accompanied by the rustling of what sounded very much like our bear bag. Getting a fire going as quickly as possible (meaning, a liter of white gas poured onto the nearest thing that looked like wood and then set ablaze), we didn't find a bear. We found a team of TWO bears attacking our bear bag.
The big one climbed up the trunk of the tree, just under the branch from which we'd hung the bag. The little one, presumably a cub of the big one, had climbed out on the branch, and in a series of small steps, had pulled the bag along the branch with one arm toward the larger bear, who could now reach it from her spot on the trunk, and who was shredding the bag to bits as all our food dropped out. The fire, of course, chased the thieving duo away after a couple of minutes, and they thankfully only got away with some sausages and most of a bottle of pancake syrup.
Of course, what we hadn't noticed was that this tree had basically no leaves or branches or bark on it anywhere. Based on the number of large scratches and claw marks all over the tree, we surmised that we weren't the first ones to try to hang our food from this tree, which was essentially a food collection station operated by the bears to tax any humans foolish enough to camp there.
The damn bears are smarter than you'd think.
He's left the kernel before (2003) (Score:2, Interesting)
http://kerneltrap.org/node/759 [kerneltrap.org]
It seems he left with little notice then (although he was maintaining the older kernel - 2.2) . Kernel development still continued in his absence....
Re:Thanks (Score:3, Interesting)
>>>The Gentoo maintainers were particularly rude
Couldn't you just avoid the forums and focus on coding? Then when it's done, hand it over and say, "Here is it. Peer review it and I'll make the changes." Ignore rude persons by killfiling them (first give them a warning; then if they are still rude, block 'em).
Re:Linus (Score:1, Interesting)
Re:Thanks (Score:5, Interesting)
Unless you're running Vista x64, have 4GB or more of memory, and are trying to install the drivers for a TV tuner. Doesn't matter which manufacturer, they all would tend to fuck up under such conditions. I gave up trying to get TV tuners to work on my rig after wasting close to $250 on various models cuz I found only one commonality amongst them: they don't fucking work.
Push-out doors bad for snow and other egress (Score:1, Interesting)
I *absolutely* agree with you when it comes to many types of doors, especially public restroom doors -- why, in the devil's briefcase, do most public restrooms have doors that you have to pull when your hands are wet from washing them? Hygienically brain dead.
Anyway, various kinds of doors that open to the outside should not generally be push-out types. Consider residential doors opening to the outside, particularly in cold climates: if there is substantial snow on the ground, the occupants will be trapped inside. Even in not-so-cold climates, push-out doors to the outside can be too easily barricaded, again trapping people inside. Plus, as others have noted, such doors perforce have the hinges on the outside, allowing too easily for forced entry.
Meanwhile, public buildings usually have push-out doors mandated by local fire codes, for precisely the reasons you note -- but they generally also have awnings or some other means of ensuring that snow build-up does not block the doors from opening.
(NB: Posting anon for modding purposes.)
Cheers,
Re:Linus (Score:4, Interesting)
Agree 100%. Something as simple as a door should not require an instruction manual, even if it is a single word ("push" or "pull"). A good designer ought to be able to design any device as simple as a door so that it is easy to use correctly and difficult to use incorrectly, even for users that are minimally observant.
I recently thought about this when I entered a bathroom stall, and the flip lever near the top of the stall door had a coat hook on it as well. When the stall door was unlocked, the coat hook was up against the stall divider and impossible to use. When it was locked from the inside, the coat hook was available for use. Furthermore, if you did decide to hang something on it such as a coat, a purse, even a key, it would take actual effort to ignore the item as you tried to exit the stall. Simple, straightforward design that is functional and eliminates the possibility of leaving an item behind.
(Unfortunately, the toilet paper dispenser gave one lone, see-thru, single-ply sheet at a time and was positioned behind your left ankle. wtf.)
Re:Thanks (Score:3, Interesting)
no one said it had to pass the resized displayed image. as you noted, it downloads the original image. why would it not pass the original image to the other application? that is the behaviour I expect, and that I've seen. and it doesn't involve my image editor going out to the web to re-get what is already on my machine, in my browser cache if not actually in memory.
Re:Thanks (Score:3, Interesting)
This was on the bug-tracker, not the forums. I've never posted in the Gentoo forums.
To be honest, I was being a bit unfair - I have the feeling a lot of the Gentoo devs are very young and quite atypical in the Linux world. A lot of the times they clearly didn't understand certain concepts.
Re:Thanks (Score:1, Interesting)
Sorry, but that wasn't my impression. All bolding mine:
Linus uses melodramatic emphasis and a condescending tone throughout the entire conversation.
I'm not in any way partial here, but I haven't suddenly been overcome by the urge to buy that man a beer.
Re:Thanks (Score:2, Interesting)
What I've learned in this Cox/Torvalds case is
Re:MS broke backward compatibility... (Score:3, Interesting)
Remember that they were also introducing a new security model, which is one of the reasons they HAD to break it. They made that choice, for better or worse.
Re:Thanks (Score:3, Interesting)
No they don't! They just have to advertise that it works, it doesn't actually have to work! This is exactly the sort of crap I went through with Hauppauge in my prior posts below. They can advertise functionality all they damn well please, and once you pay for it, bring it home and void most store return policies by unpacking the damn things, you find out that *gasp* the damn things don't work as advertised and, much later, you find out that they never will! Once they've got your money though, what do they care? You can complain all you want, but if they're not going to provide support, then they're not going to provide support, and short of a class-action lawsuit, you're screwed.
Heck, chances are you wouldn't get your money back anyway even if you won a class-action suit.
Re:Thanks (Score:4, Interesting)
You volunteer hundreds of hours of your time, without pay, and all you hear from the members is complaint-after-complaint-after-complaint.
You eventually reach a point where you say, "What am I doing this for? No one's appreciating it or saying thanks. I could be out having fun instead of this shit," and then you stop volunteering.
You just described why I mostly use commercial software. Because they take all my complaining with a smile and a nod and get to work.
Re:Thanks (Score:2, Interesting)
You just described why I mostly use commercial software. Because they take all my complaining with a smile and a nod and get to work.
Nah, they take the complaining with a smile, nod and go back to their "dev center" to laugh out loud about the whining guy who will pay another $100 for the next improved, more synergic version.
Re:Linus was right (Score:1, Interesting)
Actually the problem is more complex.
From what I understand there is an old bug in the TTY code. This, coupled with misinterpretation of what some calls (such as E_Again) should do led to programs using this bug as a feature (Alan Cox vision, which I tend to respect, especially when it comes to TTY).
Alan Cox tried to fixed the bug, but this led to breakage in some apps, notably Emacs.
A bug report was submitted. Alan Cox refused to revert his changes and started working on a solution that would both fix the bug AND allow broken apps to work again.
Linus came in and asked for a revert. Alan Cox replied that it was a bug in userland form the already buggy Emacs. Linus did not approve, and mentionned au _nicer_ fix for the problem at hand that Alan Cox refused. Alan Cox replied that Linus obviously did not understand the bug.
So Alan Cox point of view is that he corrected a bug in TTY that led userland apps that were using this bug as a feature to have problems. But that is not his problem because he is not responsible for userland. Further more he seems to think that the _nicer_ fix is inappropriate.
Linus point of view is that Alan Cox introduced a regression in TTY and that therefore he should rollback or adopt the _nicer_ fix and stop accusing userland or Emacs of being buggy.
Of course I am nowhere near the amount of skill required to know who is right. But I do know TTYs are a complex matter and that a fix that seems nice for one problem can bring thousands of problems elsewhere. My personal opinion is that Alan Cox is more apt to get the big picture than Linus when it comes to TTYs.
Re:Thanks (Score:3, Interesting)
I'm on year 11. The previous 5 years were spent working really hard, and caring about how projects at work went. When projects at work succeeded, I was ecstatic. When they failed or were cut due to outside business demands, I was mad and hurt. Then I poured my heart and soul into a project for 3 months, only to have it canceled on the day it was being rolled out. It turned out later this was a good decision, but at the time I was upset because all I could see was that they had killed my pet project. So I stopped trying to achieve satisfaction and status in life from work, and instead started treating work for what it is - a means to an end. My self-worth comes from things I do outside of paycheck hours, and I've been much, much happier.
I still try to do my best; I just don't get emotionally invested. Think of taking out the garbage or washing dishes. You need to do these things, and it's good to do them well, but you don't need to *care* about these things. You also don't have to hate them. Is it fun and exciting? No, but it's also not bad. It's just kind of there.
What's funny is that when you become emotionally detached from your work, it's easier to make decisions that would have otherwise been painful. Compromising on or sacrificing your "baby" (project) for the good of the company? No problem. Doing something that is technically atrocious but makes sense for the business? No problem. You start to think about ways to do what the business wants instead of what would bring you personally the most satisfaction - which ironically makes you a better employee. Way too many people get caught up in doing what they think is the right thing technically, regardless of what the business needs or wants. There's always the argument of what the business wants right now and what it needs long term, and you can still argue that - in fact it's your job to examine requirements and try and do what's best for the business both in the short and long term - but you don't have to be emotionally invested in the outcome.
Now for the math!
Let's assume the standard "Eight hours for work, eight hours for sleep, eight hours for what we will", so 40 hours a week of work. Not 60, because after all, we're talking about someone who doesn't care about work.
There are about 260 weekdays in a year. Subtract 12 for holidays, 25 for vacation, and you're down to 223.
There are 365*24 = 8760 hours in a year.
Sleep: 365*8 = 2920 hours (33% of total)
Waking hours = 8760-2920 = 5840
Work: 223*8 = 1784 hours (20% of total, 30% of waking)
Free time: 8760-2920-1784 = 4056 hours (46% of total, 70% of waking)
You spend more time sleeping than you do working! Even if you had no holidays or vacation, work would still only be 23% of your time and 35% of your waking.
If you sleep 7 hours a night like me:
Sleep: 365*7 = 2555 hours (29% of total)
Waking hours = 8760-2920 = 6205
Work: 223*8 = 1784 hours (20% of total, 29% of waking)
Free time: 8760-2920-1784 = 4056 hours (50% of total, 71% of waking)
So for 29% of my waking hours, I'm doing something that I neither love nor hate - and my happiness during the other 71% of my waking hours depends on what happens during those "me" hours, not on what happens in my workplace.
P.S. I wrote all of this at work. Even spreadsheets can be fun when you're using them to calculate silly things like waking hours.
Re:Linus (Score:1, Interesting)
and what does the "I" stand for, moron
Re:Thanks (Score:1, Interesting)
Just tried it, GIMP connected to the server and pulled the image from there. Not sure if that's how you want it to work though.
I'm wondering if someone does this with an image from a more...adult.....website.....
will GIMP order some hookers ?
Re:Linus (Score:1, Interesting)
and you don't think IQ is related to intelligence?
It might not be a good measure of intelligence, but it is surely related to it.
Re:Thanks (Score:3, Interesting)