This Takes The Cake: Sam Varghese of iTWire Goes On The Offensive Again

By Caitlyn Martin
April 14, 2010 | Comments: 44

Sam Varghese is a columnist for the iTWire website who likes Ubuntu and apparently has great respect for Mark Shuttleworth. He writes opinion pieces that I often disagree with but he most certainly is entitled to his opinion. He has now, on three separate occasions this year, written pieces which are harshly critical of Ubuntu's detractors. If you dare say anything negative about his favorite Linux distribution you will be in Mr. Varghese's cross hairs. For example, he took a broad swipe at anyone who didn't like the changes in Ubuntu 10.04 (Lucid Lynx) in a piece entitled Ubuntu users, Shuttleworth doesn't owe you anything. Anyone who so much as questioned the change of the position of Windows buttons is guilty of a fatal flaw common to Linux users.

Linux users possess one awful characteristic in spades - a sense of entitlement.

Nothing else can account for the way in which a number of users have tried to force their will on the Ubuntu chief, Mark Shuttleworth, and get him to reverse the decision.

Mr Varghese seems to think that Mark Shuttleworth is this benevolent philanthropist who is giving us, the Linux community, something wonderful for free and takes nothing from us. Nothing could be further from the truth. The vast majority of Ubuntu code comes from Debian, which in turn comes from the labors of dedicated members of the Linux community. In effect, for those of us who have contributed code to Linux projects that, in turn, end up in Ubuntu, Mr. Shuttleworth has taken our labor. Granted, as members of the Free and Open Source (FOSS) community we gave it willingly. However, to claim that Mr. Shuttleworth and Canonical owe nothing to the community upon which they built their business is simply not the case.

In addition, Ubuntu likes to tout the community nature of their project. The fact that Ubuntu has a large user community and generally interacts well with it says something very positive about Canonical. However, if you accept volunteer labor from that community and state that you want input from that community, as Ubuntu clearly does, you have to accept the fact that some decisions will not go down well with everyone in that community and you are likely to get some negative feedback. Canonical does not censor that feedback and, one would hope, they solicit it because community feedback helps them produce a better product. Mr. Varghese doesn't see criticism that way and instead launches on three page vitriolic attacks on the community members who dare question the wisdom of the almighty Shuttleworth.

In his pieces opposing Ubuntu's critics Mr. Varghese has been adept at spinning and outright misrepresenting the words of those he disagrees with. For example, in a February 15th piece titled I'm not driven by Microsoft hatred: Shuttleworth, Mr. Varghese so misrepresented Alan Lord's comments in The Open Sourcerer blog that Mr. Lord felt compelled to write this response, concluding with the words "Sam, your article paints me with a brush which I do not believe to be fair or accurate." Don't worry, Alan, you're hardly alone in being misrepresented by Mr. Varghese.

Indeed, in a piece published today Sam Varghese spends three pages explaining The main problem with Linux: ignorant users. However, he doesn't talk about users in general. The three pages are dedicated to attacking someone Mr. Varghese seems to see as the epitome of the ignorant user: yours truly. Oh, I've been pounced on by people who disagreed with me before. It's part of the territory when you write honest but sometimes unpopular opinions. Mr. Varghese, of all people, should know this. However, in iTWire today he treats us to a three page diatribe which does nothing but mischaracterize my words and misrepresents my opinions.

My great sin, according to Mr. Varghese:

Yesterday, I discovered one more GNU/Linux user who appears to think that Canonical does owe its users something - reliability - and is prepared to vent about it in public.
Actually, he's right to a point. If an operating system isn't reliable then what good is it? He goes on:
One would be inclined to expect better from Caitlyn Martin, who claims to have used GNU/Linux from 1998 onwards, and also advertises herself as a technical consultant with a background in several tech-related areas. But, sadly, such does not turn out to be the case.
Let's get the facts right, shall we? Actually, my bio says I've used Linux from 1995 onwards. I've worked with Linux professionally since 1998. My background includes a period in 2005 consulting for Red Hat. I have also volunteered my time and efforts packaging software for VectorLinux and helping to maintain their repository, for which I received warm thanks from their excellent development team but nothing more. As a result I do have some insight as to what it takes to develop a Linux distribution.

Mr. Varghese goes on:

Under the heading "Ubuntu is a poor standard bearer for Linux" Martin vents on a blog provided by the publisher O'Reilly, with her starting point being that she could not hot-plug a printer and get it recognised by her netbook which runs Karmic Koala, or version 9.10 of Ubuntu.

Martin wants to apparently stay on the cutting edge by constantly updating her distro but also wants everything to continue to work. She wants the same functionality that a long-term support version of Ubuntu offers to continue into the bleeding-edge releases which occur between LTS releases.

This is a string of purely false statements, the first, sadly, of many in Mr. Varghese's piece. I did refer to a regression in either udev or devicekit which caused my HP LaserJet 1020 printer to fail to work if it was on prior to booting. In other words, the failure is in cold-plugging, not hot-plugging as Mr. Varghese states. I should add that this regression impacts all printers which use foo2zjs, a driver for printers that use the Zenographics ZjStream wire protocol, which includes models by HP, Konica and Minolta.

This printer worked fine under Ubuntu Hardy Heron (8.04 LTS) but failed after I upgraded to Karmic Koala (9.10). You will note that these releases are not six months apart but rather 18 months apart. I specifically do NOT recommend upgrading every six months but rather when necessary either because of a lack of support for new hardware or because a release is going out of support. To claim that I want to upgrade every six months is utterly inconsistent with both what I wrote and what I've done with my own systems.

His next point, that I want six month standard releases to have the same stability as LTS releases is also inconsistent with what I wrote. What I did say was that Ubuntu sells itself as "Linux for human beings" and Canonical press releases tout each and every release as stable and an improvement over all that has come before. In my second piece (which Mr. Varghese terms a "rant") I stated:

There are a number of ways that the issues I raise can be tackled. One would be a simple marketing change: tout LTS as the stable "Linux for human beings" and the six month releases as cutting edge. In other words, Canonical can change the expectations of the user community without substantially changing the product and then meet those expectations.
In other words, I am not demanding or even asking for a change in the Ubuntu product. What I am asking is that Canonical live up to the promises it makes to the community and to the wider computer using public. What I am asking for is no more than truth in advertising. I did offer up improved QA on the six month releases as an alternative solution if Canonical should wish to make that choice.

Mr. Varghese continues:

Her main grouse appears to be the fact that Ubuntu is now recognised as being more or less synonymous with Linux; other distributions, she claims, are better. Have the good folk at Canonical done something wrong by marketing their product well? Is it anybody's fault that Ubuntu head honcho Mark Shuttleworth appears to have a few spare coppers that he wants to spend on developing and marketing a GNU/Linux distribution?
For crying out loud, when did I ever say any such thing? It is neither in either of my articles nor in my responses to the comments. My complaint isn't good marketing. It's misleading marketing that gives false expectations to newcomers to Linux. When a newcomer to Linux is promised a simple, problem free computing experience and fails to get anything remotely close to it they walk away with a decidedly negative opinion of Linux as a whole, not just Ubuntu. This wouldn't be the case if Ubuntu hadn't become all but synonymous with Linux to the general public. That was the point of my original article.

The follow-up piece, the one Mr. Varghese calls a "rant", was my admission that I didn't express my ideas as clearly as I should have done in the first piece. It wasn't a rant and it did make clear that I think that Canonical and Ubuntu do some things incredibly well, such as getting systems preloaded with Linux into the marketplace. I also commended Ubuntu engineers for making certain those systems work flawlessly, as was the case with my HP Mini 110 netbook, delivered with Ubuntu 8.04 LTS installed. Further, I acknowledged that is easier to do with a limited hardware set than with a general distribution. Of course, Mr. Varghese ignored this part of the article because it didn't fit in with his desire to paint me as an "ignorant user" who has a "sense of entitlement."

Mr. Varghese asks:

Or does it irk her that, horror of horrors, the man is actually trying to make money off Linux?
Say what? I thought I was the person who insisted that having corporate or institutional backing for a Linux distribution is a positive thing as it insures the continuity of the product.

Of course, Mr. Varghese was just warming up:

There are so many contradictions in her rant that anyone who takes her seriously would have to be smoking some rather strong stuff. For instance, at one point she writes: "There has been a lot of discussion about why Ubuntu consistently fails to deliver a stable, reliable product."

But the very next line reads: "Ubuntu 8.04 LTS (Hardy Heron) was, in fact, very stable and reliable after the first maintenance release (8.04.1) so I have no doubt that Canonical can produce a quality product."

What contradiction? The initial release of Ubuntu 8.04 LTS was buggy and had significant problems. Even some of the most strident defenders of Ubuntu who responded to my pieces admitted that much. However, much unlike subsequent releases, Ubuntu fixed 8.04. I quoted and linked articles by the likes of Andrew Wyatt, Chris Smart (from Linux Magazine, no less), Carla Schroder and Akkana Peck that demonstrated an interminable series of regressions and severe problems with every release since 7.04 (Feisty Fawn). Are those who take me seriously "smoking strong stuff" or is Mr. Varghese suffering from a basic lack of reading comprehension?

Mr. Varghese also conveniently forgets that my main complaint is not that Ubuntu has bugs. Every Linux distribution has bugs. Every operating system release of any sort has bugs. What really irks me about Ubuntu is they have more serious regressions and bugs that affect more users than are typically found in a Linux distribution. Even that wouldn't bother me if they managed expectations for their cutting edge releases the way that Fedora does. No, what really bothers me, and what I expounded on at length, is the fact that Ubuntu chooses not to fix those bugs during a release cycle while claiming that each release has 18 months of support. If you don't fix problems are you really offering support?

From here Mr. Varghese devolved into calling me "stupid", "ignorant", a "spinmeister", someone who doesn't read what she writes, and on and on. Indeed, in the entire three page attack Mr. Varghese doesn't once refute anything I've written with facts to the contrary, nor does he provide any evidence whatsoever which indicates that I've written anything which isn't completely accurate.

Mr. Varghese, I wrote my own piece back in November stating what I believe is the number one problem with the Linux community. What I didn't do was engage in ad hominem attacks on specific people or misrepresent what they wrote. I did expect better from a Linux columnist from a well regarded website. I see I was wrong to do so.

Mr. Varghese is correct in one of his criticisms leveled at me: there is nothing new in my articles about Ubuntu published this week. I am reporting a continuation of the same old problems with Ubuntu that have been detailed by writers who are far more prominent than I am over the last three years. All I am doing is pointing out that nothing has changed. Indeed, by failing to fix bugs and answer complaints Canonical and Ubuntu are showing a disdain for their user community that I expect from some proprietary software companies, especially Microsoft. The fact is that I hold Canonical in higher regard than that and I, and many, many others in the Linux community, expect more from a Linux distribution. I see nothing wrong in that.

As I stated at the outset, I do think the Linux community is entitled to some things considering that a significant percentage of us have given back to that community. Canonical and Ubuntu have benefited directly from community efforts in the same way most every other Linux distribution has. As such they do have a degree of accountability to that community, even those of us who have not made direct contributions to Ubuntu.


You might also be interested in:

44 Comments

He's long proven himself a deluded twit, this just reinforces that view. As you said, there is nothing "new" in what you are saying, but you have indeed laid it out far more clearly and succinctly than many others (so far as I am aware), and have coped quite well with the childish and immature venom directed your way by the "fanbois", of which I feel was far, far out of proportion, and completely childish. You have definitely cememnted my respect for you with the way you have carried yourself thru this "ordeal" Ms martin, and salute you for doing what you feel is right, and "sticking to your guns". I have faith in you, and shall be now paying more than cursory attention as have done in the past. Thank you for being you. Cheers. :)

David

All I have to say is that Sam comes off looking like a tool with his article. Hopefully, he's man enough to apologize. If not, the little jealous boy in him wins and a pony dies.

Thanks for the clarification Caitlyn

devnet: to my knowledge, Sam's never apologized to anyone for any of his wildly inaccurate personal attacks.

I really don't know how the guy still has a job, at this point.

Mr. Varghese hasn't even corrected the factual errors I pointed out, i.e.: that the problem with printing refers to cold-plugging, not hot-plugging, or that I've used Linux since 1995 rather than 1998. It seems accuracy isn't of great concern to Mr. Varghese.

Sam Varghese, like the "Boycott Boys" from BoycottNovell, are never interested in the facts, but simply making vicious personal attacks.

Amusingly, Sam got upset when someone attacked Roy for his lies/mistruths under the guise that it's not right/unfair/whatever to attack a "poor college student". A few months later, Sam writes a 5-page personal attack against a college student in India that vented his frustration about the GPL, trying to claim that the kid was Miguel de Icaza's "right-hand man". By the time Sam had even published the article, the kid had taken down his rant and apologized and Miguel didn't even know who the kid was, nevermind being his "right-hand man". (I guess Sam figured he could smear two birds with one stone?)

Sam Varghese has an agenda, get in his way (or, rather, disagree with his world view), and he will smear you with everything he can.

He's done it countless times before (Miguel de Icaza, Jeff Waugh, that Indian college student, Matt Zimmerman, etc etc etc) and he'll continue to do it. Same with the BoycottNovell crowd.

(Notice how Sam and BoycottNovell try to smear various Canonical employees now because of some conspiracy theory about how they are Microsoft plants?)

These people are poisonous to the Free Software community.

I'm surprised at all the attacks that've been launched against you in the last few days, caitlyn. The way I see it, whether or not we as individuals have contributed code back to Linux, we're all at least entitled to a distribution that works as well as Shuttleworth claims it does.

At any rate, at least Varghese didn't claim he wasn't going to "feed the troll" (you) and then launch a personal attack against you on his blog, like some other moron did a day or to ago.

I think he did a splendid job of representing what's wrong with the Linux community .. people like himself.

One thing that isn't said enough is that you don't have to contribute code to be a member of the community.

Anyone that has ever helped out in a forum, spent time in a LUG helping people, submitted an idea, or has just answered a question is just as valuable to the community as one that submits patches.

The next time someone tells you to patch it yourself, just remind them of that. :D

*humps Fewt's leg* :P

Ewwwww...

Thanks for posting that. There are myriad ways non-technical people can contribute to their favorite distro, ranging from artwork and graphics to translating documentation. Providing community support is an essential part of the Linux ecosystem. In the case of small, community-based distros the less than glamorous tasks only get done when someone steps forward and offers to help. Everyone has skills that can be helpful and nobody should feel they have nothing to contribute. Thank you, Fewt, for an excellent reminder.

I saw that article and it is ridiculous that he wrote so much that it took 3 pages to fill just so that he could attack one individuals opinion. Actually I like Ubuntu but it does not fill my needs as well as other distributions of Linux. What is wrong with reminding the world that there are other options and that Ubuntu needs to improve. I hope they make needed improvements with the Ubuntu system I like it and would like to be able to use it more.

Hah, by responding to him you're feeding the Troll. Don't feed the troll.

I did glance through his article and other than choosing to attack you personally and getting some facts wrong I thought his points were legit though. Some of your articles definitely have that sense of entitlement seeping through them, your comments about Microsoft are silly too. I understand you don't like MS products but to label their products as "crap down your throat" makes you a bit of a troll too. For all the things Linux does well it has a long way to go until it establishes the userbase Microsoft products have. Anyways, I digress, preach on with your stuff, I consider both of the sides of this argument as soap.

@Niva: Even though Mr. Varghese's article is dated today he actually posted it about 48, correction: 36 hours ago. I thought long and hard about whether to respond or whether I would be, as you say, feeding the troll. There was one blogger who wrote a piece called "Caitlyn Martin is a troll!" in response to what I said about Ubuntu. I thought it was stupid and not worth responding to. What makes Mr. Varghese different is that he is an editor for iTWire. Just as the O'Reilly name gives me an extra degree of credibility, he derives a level of credibility due to his position with a widely respected tech website. He isn't a garden variety troll. He isn't just another blogger doing a rant. That's why I decided to respond. I did consider the very point you make, though, and it is fair comment.

Regarding Microsoft products, I just completed a five month contract where, for the first time in quite a few years, supporting Windows was a significant piece of my responsibilities. "Crap" is a mild word for what I think of Windows. Let's put it this way: 50% of the time spent by the desktop support folks is dealing with malware. That's behind a firewall, mind you, with Windows machines that, for the most part, are reasonably well, by-the-book secured. They pay for support for one of the leading anti-virus products and it catches an awful lot but even with automated updates of all the latest virus signatures and such enough malware still gets through to make desktop support hell.

What really galls me is that there is nothing that organization is doing in Windows that couldn't be done in Linux. They don't have specialized software that would be impossible to replace nor do they have custom apps that would have to be rewritten. In theory they would be an ideal candidate for Linux on the enterprise desktop. To make matters even more ridiculous, they are located in Raleigh, the home of Red Hat.

In regards to the "down your throat" part I'll ask if you've ever read Microsoft enterprise contracts. If not I'd ask you to withhold your criticism until you do.

Sam Varghese's favorite distro is Debian Stable. If you read his older articles and forum comments you will find he has been a long time critic of Ubuntu. His recent defence of Ubuntu is a bit of a laugh for those of us that have been reading his stuff for years.

Now in defence of Ubuntu .....

I am currently using Kubuntu 10.04 beta 2 on my desktop and Ubuntu 10.04 on my notebook, both are working flawlessly. My Canon MP640 works beautifully over my home network. My TV cards (internal and USB), FM radio card, GPS dongle, my daughter's iPod Touch and my own Creative Zen 32gb are all working correctly. Add to that wifi and my 3G Mobile internet dongle. Yes, you can add/remove songs from the Touch using Rhythmbox. All of my hardware is purchased after I research Linux compatibility, only an idiot buys a piece of equipment and then whines that it does not work. In our house we also have a midi keyboard running on an Ubuntu-Studio box and a Media PC driving a TV with optical spdif to an external Yamaha amp for 5.1 suuround sound, networked and, of course, running Ubuntu.

I have used every distro under the sun, the BSDs and even ran Gentoo for a year compiling everything by hand. I have Sidux on a number two desktop, just because I can. Yet I run Ubuntu on my main computer because, most of the time, it just works - no mucking around. Five minutes with Google before buying any new hardware saves so many hassles, tailor your computer to the OS (that is what Apple does after all).


I would say you've been exceptionally lucky. While I've never said that Ubuntu doesn't work well, even flawlessly, for some people I've become convinced that there are issues for a very sizable percentage of Ubuntu users. There are a lot of factors that impact whether or not a bug or regression will hit you: the hardware you run, the apps you run, your usage patterns and so forth. All the articles I linked and all the bugs I mentioned are well documented, including in the Ubuntu forum and on Launchpad. The problems are very real. I'm glad they haven't affected you at all.

I haven't tried Lucid yet. I just don't have the time to be a beta tester now and what I have (SalixOS) works. I almost certainly will have to at least look at 10.04 for a bit after release. I honestly hope that Canonical has a superb release to tout this time. It would be much better for the Linux community as a whole, not to mention their users, if they do.

In other words, I don't discount your experience at all. I just wish more people had that experience. I also wish Ubuntu would fix bugs within a release cycle. They will with Lucid since it's an LTS release and they are trying very hard to attract enterprise customers. I just wish that they would either stop touting the six month releases as stable and ready for the masses or else that they do what every other major distro does and fix the bugs. Despite what Mr. Varghese wrote I really am not asking for anything at all out of the ordinary.

Caitlyn

I have often thought that Ubuntu should label each new release as testing, and then declare them stable when the next release arrives. Thus 9.10 would become the stable release at the end of this month. If nothing else it would manage the expectations of the users and allow 6 months of development and six months of usage (and bug fixing) before being considered "stable". Then there would be three Ubuntus - testing, stable and LTS. The problem with any LTS is that latest hardware support is only ever in the latest kernel - not just Ubuntu but all Linuxes, so a two year old release is two years behind for hardware.

Actually, the latest hardware support is not only in the latest kernel. Red Hat famously backports new hardware support into their stable kernel for Red Hat Enterprise Linux. This result has produced incredibly solid and reliable code. Since CentOS and Scientific Linux use the Red Hat kernel they, too, provide new hardware support in older kernels.

"Five minutes with Google before buying any new hardware saves so many hassles, tailor your computer to the OS (that is what Apple does after all)."
What about newcomers from the emerging countries (which contributed to a looong waited increase in "maket" shares last two years?) Are you sure :
a) they know what google search is?
b) they have the means of google-searching before buying their first computer (and, if it is a linux one, will be upset if there is bad support)

If you stick to mainstream video cards (not the absolute newest and expensive), HP, Epson or Canon printers then Linux will "just work" with built in drivers or readily available drivers. Digital cameras are 99% plug in and use. Of course some clown will pipe up that their old "X" model does not work, get over it, it is a tiny proportion of the equipment available.

The older and simpler the computer gear the more likely it will work with Linux. There is plenty of older gear that will not function under Vista or Windows 7, but will with Linux - because Linux tends to keep drivers forever. Linux has drivers for the vast majority of hardware, drivers built into the system. Sure there are obscure printers or USB rocket launchers that may not work, but then there are also plenty of such devices laying idle because no one will write a new Windows driver for rare or older equipment.

The best thing about Linux is that it is free, so I will not cost the third world person anything to try a Live Linux CD on the gear that they are thinking of buying. And are you really suggesting that there is anyone in the world outside of a desert or jungle that has not heard of Google, even if they do not have internet access. Those few people who have not heard of Google would also be the ones who will not be buying a computer either.

Thanks for your pavlovian mantra, and I will quote (small thx Dog) pieces of it?

"If you stick to mainstream video cards (not the absolute newest and expensive), HP, Epson or Canon printers then Linux will "just work" with built in drivers or readily available drivers. . "

But do people who buy ****cheap*** netbooks -they are new and cheap!, that exists , in the real world?- stick to mainstream video cards?
FYI : this was the only part of the PC market which has been significantly growing, part of them with some kind of GNUlinux preinstalled, thus contributing to theses 2 last years increase in GNUlinux popularity...

And why, oh why was the Intel driver bug introduced by Mandriva and UBUlinux (Fedora, Sabayon and slackware provided decent response times to a mouse click, not a matter of *tens* of seconds-)?
Mandriva fixed it, but it might be insufficient for absolute newcomers,
UBUlinux pissed off people asking for fixes (and sometimes even providing a fix!)

Does a looooog credo provide serious explanations?

"Of course some clown will pipe up that their old "X" model does not work, get over it, it is a tiny proportion of the equipment available."


Where the netbooks owners, when they suffered from this bug (which made Internet interactions terrible)
"clowns"????

" And are you really suggesting that there is anyone in the world outside of a desert or jungle that has not heard of Google, even if they do not have internet access. Those few people who have not heard of Google would also be the ones who will not be buying a computer either.
"
Do you know what is the average speed of the Internet in a developped country ( i.e Europe, Canada, Australia,
not in a jungle, nor in a desert, though I know the speed in Brasilian _savannas_)?

Give figures, please, then!!!

And what the price it is to wire the data cable, in a developped country, not in the caricatural desert_and_jungle?
And lots of people buy computers (and bought, before THE Internet existed) to text process, develop,
even to compute!
and do not need google at all (or marginally : the reflex to go to the fora to whine for help is limited to a small part of the world)..

Caitlin, it is a shame that such people exist in today's world just to demean and belittle someone without having all the facts.
I saw that today first hand when a devout Windows user was accusing all Linux users of being hackers, sticking viruses and malware into Windows. I set him straight.
You're absolutely right about Ubuntu. When someone produces a Linux distro it should live up to the users' expectation, which Ubuntu failed to do.
I am 100 percent behind you. Thanks for writing the truth about Ubuntu. This disabled military veteran and fairly new Linux user appreciates it.

So you're Locustus? If so, I must say your article today about that hacker was both funny and a sad commentary on the Windows user in question. I enjoyed reading it. You've written some wonderful stuff.

Honesty and integrity are values I admire most about our various Linux communities.

Saying something is broken could be the first step towards fixing/improving it. If we all shut up and didn't complain we'd be spending more time modprobe-ing and less time working.

Don't lose faith Caitlyn. The truth needs to be told.

QUOTE: What really irks me about Ubuntu is they have more serious regressions and bugs that affect more users than are typically found in a Linux distribution. END-QUOTE

Yup... And for a linux touted as "for the masses", an experienced linux person and devout distro-hopper with 10 years experience lost access to several other installed distros in my box because Ubuntu's install program kidnapped my grub and installed its own grub without any prompt or warning whatsoever.

EVERY other linux install I have done has given me the option to tailor grub in some fashion or another, and most give me the option not to install grub if I don't want to.

I am NOT a linux guru, but I have had consistent troubles with ubuntu/kubuntu since first experimenting with it in 2005. At that time, I devoted a great deal of effort and time into kubuntu, but the experience was fraught with difficulties.

Ubuntu does some things well, but it falls short in many ways. Just like most distros.

So, your points are very strong (that ubuntu shouldn't tout new releases as stable, but should tell users that the new releases are "testing" and give some indication about the most basic of things, that there standard install CD will kidnap your present grub and install over it).

D.

I have to agree on all points in this and the last two articles.

I had a nice Avaratec ultralight which was like a netbook, which had a VIA chipset. Edgy worked. Every version of Ubuntu past that got worse to the point where I only have 800x600 VESA today. I forgot where, but they removed the tool that let you manually specify the resolution. The LCD was 1024x768, and I used some GTK app and saved the config and was done with it.

So why do I still use it? The only reason is they have a large number of repositories so I can obtain packages for a lot of special utilities I use, including things like medibuntu (one radio station stream uses a proprietary codec, and I do like GoogleEarth, though I should switch to open streetmap). But simply, it makes it the most adaptable system. Debian tends to lag a bit, and technology changes too fast.

The last go-around was with bluetooth PINS. It was mainly Gnome who wanted to create a huge database of default pins, but I pointed out there were too many, submitted a patch that was improved and eventually incorporated.

I'm using Xubuntu since it is smaller (I can add if I need things), is still configurable, and far less annoying.

I shouldn't but if you will indulge my set of longstanding complaints:

Network-Manager Applet doesn't have a refresh button (other than disable/reenable wireless), but can take several minutes to add to the menu or connect to an AP where "sudo iwlist scan" will show it in the list immediately. I blame Gnome for being stubborn in insisting that "refresh" is so much harder for the user than waiting 2-3 minutes, but I blame Canonical for not accepting a patch to fix it. Again for TWO YEARS.

notify-osd, polkit-1, and grub-2 are broken. They were in Karmic. They still are in Lucid. Notification had more utility, but the replacement is annoying and completely non-configurable (I uninstall it first thing). It also doesn't go away (too slow?) when I hover. polkit-1 has URLs for authentication which I think are supposed to let you set the default (authenticate for session, forever, for everyone, for just you), but I end up editing an XML file somewhere in /usr/share to fix it. Things like mounting USB keys. GRUB-2 can't find bootable DOS partitons, but will find any junk on USB keys or other removeable drives and add it to a bedsheet menu. Including MacOS which won't/can't boot.

These are 3 sigma items for me and the rest generally works (and I'm getting skilled at neutering the latest release), but for your printer problem there is probably someone out there who is in the same position as I am - and the forums have fixes for all the common problems, often for years, but they never make it into the mainstream.

These are serious problems/annoyances. At some point they have to integrate new things. But they also need to fix them in the very next (6 months?) release or roll back the change. Don't tell me they don't have the resources to add preferences and fix the complaints about notify-osd. Or get polkit-1 to work. Grub-2 might be too big, but they at least should trim it so it defaults to only adding local drives to the menu.

Worse, I could probably fix any or all of these problems. I go back to Yggdrasil Linux with a 0.9x kernel, I have an older SCSI driver and the bogomips calibration I think is still my original speedup. One of the older Xorg LCD chip drivers has my code.

I believe that if I handed them fully tested patches completely fixing these annoyances but preserving what they want on a silver platter and volunteered to support and (unlike the rest, actually fix) any issues for the next 6 months, they would still reject the patches.

After my experience in trying to get the broken Bluetooth PIN fixed, something short and simple - well - Shuttleworth is doing a Steve Jobs imitation and Launchpad and repositories the iTunes App Store.

And why bother testing alphas or betas? They won't fix anything I find whether I report it alone or actually research and provide a fix. I have no idea why they even bother with alphas and betas if they won't fix things before the release.

I want Ubuntu fixed. I could fix those sections of Ubuntu that I've found broken. I could even do so in such a way that Shuttleworth's arbitrary style decisions would all default to current behavior, and only in the cases where the user might want (e.g. on an HD screen, notify OSD is too small, on a netbook it is too large, and it should be dismissable as an option).

But for some reason they want it broken. They want it annoying. They don't want to backport fixes. I don't know why. It isn't CANTFIX, but WONTFIX. And even WONTletUFIX. But the window decoration/controls will be locked on the left.

"I had a nice Avaratec ultralight which was like a netbook, which had a VIA chipset."

Have you tried creating an xorg.conf and loading the OpenChrome driver? I have configured a mini-itx board with a Via Chrome chipset and Lucid works well with 2d acceleration and xvideo is fully functional. It drives a 1280x1024 panel at the correct resolution. In my experience automatic video config will not work on a Chrome chipset irrespective of the distro - it is an xorg thing.

Try VectorLinux 6.0. It correctly and automatically installed the OpenChrome drivers on the Sylvania g netbook I used to own and configured them correctly. If itty, bitty VectorLinux can get it right then surely Ubuntu should be able to.

SAM VARGHESE: MICROSOFT WHORE?

This Sam Varghese appears to be a British nutcase and deranged author. I quit reading his blather about "coppers" half way through because it's boring. He appears to have a hatred of Ubuntu and is possibly a Microsoft whore journalist? Microsoft has a $50 billion checking account and they use it.

@Raji Ramsharma

HaHa

Sam is an Indian-Australian and devout Debian user. If Microsoft ever paid him anything they would surely want their money back.

That is quite correct. One thing I deliberately did not do is attack Mr. Varghese personally the way he did to me. I'd appreciate if my readers would do the same. Let's take the high road here and stay out of the gutter.

My EeePC 900 netbook works flawlessly with Ubuntu NBR 9.10. Video, wireless, function keys - everything. It also worked perfectly well with Debian Squeeze.
My Dell 11z notebook works flawlessly with Ubuntu 10.04 beta 2. Both use Ethernet, wireless N with wpa2 encryption and 3G USB dongle.

My daughter's Dell Inspiron Mini 12 took a lot of work to make Ubuntu run, because of the GMA500 driver issue - an issue that affects all Linux not just Ubuntu.

The Intel video driver regression you refer to was just as much a fault of the intel xorg devs, and it never affected day to day use on my EeePC. Using ext2 on my SSD sped up the computer much more than any video issue. For me, Ext3 caused delays in things like typing because the SSD did not function properly with journalizing. With ext2 the EeePC always worked and it was always usable even with the worst version of the Intel driver.

Any one could have added the appropriate PPA repository and got the latest Intel drivers, but that does require research and a willingness to help yourself.

Any computer needs work after the initial OS install. Windows takes hours with driver disks and reboots just to get a working desktop and a few applications, and if you have lost a driver disk then there is the minefield of dubious driver sites, driver versions and OS reactivation. With any Linux install updates are essential and so is installing such things as firmware. Some devices (a small percentage) require manual tweaking, like the VIA Chrome video drivers mentioned above. It takes work and patience. The only other option is to buy a computer with the OS (Windows or Linux) already installed and take what you are given.

If you do not like Ubuntu or it does not work for you then don't use it. Feel free to use Fedora or Mandriva or any of the hundreds of others. I do not mind.

If you want rock solid get Debian Lenny and forget the rest.

"The Intel video driver regression you refer to was just as much a fault of the intel xorg devs," You are into the territory of an article I am about to submit to DistroWatch, but without giving away too much let me say that I completely and utterly disagree with you. Distros as diverse as Fedora, Slackware, Pardus and VectorLinux all avoided the Intel driver regression. Slackware, which is really aimed at advanced users, decided to include four different Intel drivers with their distro, including substantially older drivers that worked well for most people. VectorLinux only shipped the older, proven driver and X.org at the expense of not supporting the latest hardware. Pardus shipped patched, beta code which resolved the issues for about 90% of users, and Fedora, which released later, had to substantially delay their release due to installer issues which gave them more time to get a working Intel driver into place.

My point: the distributor bares sole responsibility for what they include in their distribution. No distro is under any obligation to ship with broken code. Period.

"it never affected day to day use on my EeePC." So what? That statement would be relevant if Ubuntu was a distro aimed only at EeePC users. It isn't. The number of users impacted by this regression was simply huge. Mandriva also ended up including a fix in their repository so that eventually 2009.1 did work on hardware with the Intel graphics chips. Ubuntu NEVER did for Jaunty Jackalope and there is still no excuse for that.

"For me, Ext3 caused delays in things like typing because the SSD did not function properly with journalizing. With ext2 the EeePC always worked and it was always usable even with the worst version of the Intel driver."

Irrelevant, besides the point, and really a poor comparison in any case. Ubuntu shipped with ext2. You had a fix for that issue. There was NO fix for the Intel driver issue. With the Sylvania netbook I had at the time I was simply out of luck.

"Any one could have added the appropriate PPA repository and got the latest Intel drivers, but that does require research and a willingness to help yourself. "

First, PPA (developer) code is not supported by Canonical so that isn't a valid solution. It's a third party fix. Second, Ubuntu claims to be "Linux for human beings'" They deliberately target non-technical, mass market users. What you suggest is way beyond such people. It would be a valid solution for a distro that doesn't make the claims Ubuntu does if, and only if, the patches were supported.

"Windows takes hours with driver disks and reboots just to get a working desktop"

Absolutely. It's another meaningless and irrelevant point. What percentage of ordinary Windows users ever install the OS? Until Ubuntu is widely available preloaded in stores they have to assume that most users have to do the install. I would also argue that Linux comparisons to Windows are pointless for a wide variety of reasons. Windows is the incumbent. Linux is the upstart and has to be substantially better in many ways to even be considered by Windows users. It has to meet the claims it makes. Claiming that Ubuntu is ready for mass consumption when things are severely broken and not fixed as a matter of policy by Canonical amounts to false advertising. That is my main point which you utterly seem to miss.

"If you do not like Ubuntu or it does not work for you then don't use it." What I like or dislike is utterly irrelevant and has nothing to do with the issue at hand. The issue is that by successfully, through marketing and hype, the name Ubuntu and Linux have become interchangeable to people outside the Linux community. By continually offering a severely flawed product and claiming it's "Linux for human beings" Ubuntu damages the reputation and impression of Linux as a whole. Your whole argument had absolutely nothing to do with what I am saying in the two articles and is taking specific examples of problems and trying to deflect responsibility.

"If you want rock solid get Debian Lenny and forget the rest. "

Nonsense. Explain to me how Debian Lenny is more solid than Slackware or Red Hat.

"My EeePC 900 netbook works flawlessly with Ubuntu NBR 9.10. Video, wireless, function keys - everything. It also worked perfectly well with Debian Squeeze."

Not on release day it didn't. It kernel paniced when you pressed FN-F2, and you had terrible video driver problems. I'm sure you updated from a PPA to the safe driver upgrade, or you likely found our article in the Eeebuntu forum on how to fix it. I personally know of thousands of Eee PC users impacted by this, so stop lying.

"My Dell 11z notebook works flawlessly with Ubuntu 10.04 beta 2. Both use Ethernet, wireless N with wpa2 encryption and 3G USB dongle."

Good for you.

"My daughter's Dell Inspiron Mini 12 took a lot of work to make Ubuntu run, because of the GMA500 driver issue - an issue that affects all Linux not just Ubuntu."

Think back to this, and imagine a consumer trying to make this work, what would happen? Mark has said point blank that consumers are his target.

"The Intel video driver regression you refer to was just as much a fault of the intel xorg devs, and it never affected day to day use on my EeePC."

No no, it was the fault of Canonical for releasing this driver, when there was a safer option (proven by the safe driver PPA).

"Using ext2 on my SSD sped up the computer much more than any video issue. For me, Ext3 caused delays in things like typing because the SSD did not function properly with journalizing. With ext2 the EeePC always worked and it was always usable even with the worst version of the Intel driver."

This is a load of crap, and proves that you don't really understand your issues. Removing the journal increased the life of your SSD but it functioned fine. The only performance impact is caused by atime, an issue which consequently you still have with ext2 (change your mount to noatime or relatime).

"Any one could have added the appropriate PPA repository and got the latest Intel drivers, but that does require research and a willingness to help yourself."

A consumer cannot do this, and it is not supported by Canonical.

"Any computer needs work after the initial OS install. Windows takes hours with driver disks and reboots just to get a working desktop and a few applications, and if you have lost a driver disk then there is the minefield of dubious driver sites, driver versions and OS reactivation."

Windows and Linux take roughly the same amount of time. The only improvement with a Linux installation is that 60% of the software you need is already installed.

"Windows and Linux take roughly the same amount of time. The only improvement with a Linux installation is that 60% of the software you need is already installed."

Well, I must disagree with your last sentence as I suppose it is about install times; those install times are very chaotic;
for instance, once I gave a CD with OpenOffice (for Windows 32 bits) on it for a student; I knew it would take me 5 seconds of reflection (not of baby sitting) to have it installed, at the place I want to, with my favorite locales; I asked if it worked and was answered *and swored* the student was helped by two of her colleagues and it took three hours....

I know installing Mandriva often takes 15 minutes of work (not of baby sitting : I do something else while it formats a partition, say) to me -and I do not claim to be competent-(which is very different of one to three hours for XP, by a very competent cybercafe landlord). But it is more complicated than installing OOffice_setup.exe!

Therefore, installing a linux distribution might take weeks to the masses Ubingtu is meant for (installing XP in Europe is very fast: the graphical installer requires signing a check; the CLI installer requires typing a credit card number on a keyboard).

If regressions/bugs are added, masses will get desesperate :
the wild world is so hectic they wonot find neighbors to help them (with XP, it might happen), and if their screen freezes on their onely computer, the help from the competent Internet, full of GregE's clones ready to give hundreds of lines of reliable advises, could not even be reached.
If they get desesperate and feel swindled, I suppose they wonot recommand GNUlinux to their friends/family/bosses..

"The Intel video driver regression you refer to was just as much a fault of the intel xorg devs, and it never affected day to day use on my EeePC."

But, though it was upstream's fault (oh, the naughty one one hides behind!), Mandriva managed to have it corrected; all the other distributions did not ship it (ex : Wolvix ****beta****!)
"My daughter's Dell Inspiron Mini 12 took a lot of work t"
Did your daughter do the work?
Are you the father of every eeeeeeeePC user?

"Any one could have added the appropriate PPA repository and got the latest Intel drivers, but that does require research and a willingness to help yourself.

"
Oh, that is really the audience "net""books" are meant for!
And every "net""book" buyer/users has the (very small) skills needed, does not (s)he?

If these skills are not that widely spread (or poorly explained) , and if other distribution managed to fix this bug (or to test before distributing, why not? just a matter of common sense)
is Ubingtu ready for the masses, as it is claimed (perhaps the ignorant masses are ready for Scientific Linux, surely they are ready for Windows)

My last post here.

I have offered my experiences with the difficulties you all are so keen to attack. Nothing was without solution for a person willing to help themselves.

I did not set out to be the defender of Ubuntu in this forum - it has been forced upon me. I often switch between Debian, Sidux and Ubuntu, at the moment I am using Ubuntu and ON MY EQUIPMENT IT ALL WORKS.

It is simple - if you do not like Ubuntu then do not use it. If some piece of hardware does not work seek a solution or switch distros. When buying hardware, seek Linux friendly equipment.

@General Motors - you tell me any distro that will work better on a Dell Mini 12. Fedora and others only have hacked versions of the driver supplied by Dell for Ubuntu 8.04. That fault is %100 Intel for using a GMA500 chipset that has restrictions on source code distribution. The Dell Mini 12 is not Linux friendly, any distro - and Dell not longer sell them.

@Caitlyn - Ubuntu still worked with Intel chipsets, just with reduced performance compared to previous versions. You imply that millions of users were left with video acceleration and that is not true. "So what? That statement would be relevant if Ubuntu was a distro aimed only at EeePC users. It isn't." On my low powered EeePC with a Celeron 900Mhz chip it was still perfectly usable - that is so what. On another desktop machine I had at the time with integrated GMA3100 the reduced performance was not even noticeable.

And finally - you are all kidding yourselves to suggest the masses use Red Hat Enterprise - the Enterprise bit should give you the clue.

You still utterly miss the point and put words in my mouth freely. Yes, on your equipment it all works. As I explained in the article "it works for me" is NOT an answer. It doesn't help all the people it doesn't work for.

I NEVER suggested the masses use Red Hat. Where on earth did I say such a thing?

Once again, my personal tastes, like yours, are utterly irrelevant. What I am writing about is Ubuntu's impact on the wider Linux community and on Linux adoption. If you can't understand that, I'm sorry, but I can't help.

The Intel issue on many of the chipsets made Ubuntu utterly unusable. If by reduced performance you include lock ups, video artifacts and slowness to the point that you could have a nice cup of tea waiting for your system then I suppose you are right. My main point, which you again ignore, is not only did Ubuntu ship with code not ready for prime time but then they refused to patch it.

Ubuntu worked on your EeePC. Once again: so what? You had good results. Does that discount the experiences of all those who had problems? Should all distros be judged solely on how the work for GregE?

Once again, your answers either entirely miss the point or are utterly beside the point.

Looks like the last post was the penultimate post

If you are going to write such an article then a big question to answer is "what percentage of Ubuntu users were left in the lurch?" And you cannot answer that because only the people with problems make forum posts. And compared to other distros the Ubuntu forums are mammoth. What if we are talking about a half of a percent of Intel users? What percentage of total Ubuntu users would that be? Attack my answers all you want, my point was that it worked for me on several machines, including a very low power one, and I bet it also worked "good enough" for MILLIONS of others. I already know that you will disagree with that statement. Should Ubuntu have fixed the problem? Yes, as long as the fix did not involve a wholesale mini version upgrade that would disrupt millions of users. For the sake of how many? Saying other minor distros fixed the issue is no reason to think it would have been easy for Ubuntu, given it's huge user base. The fix involved kernel, xorg versions and driver versions - the fundamental system. I would hope you are planning to interview Jono Bacon and ask for Ubuntu's side of the issue.

Ask yourself why Ubuntu is used by millions, if not tens of millions, of users. That is the impact Ubuntu has on the wider community. Why do Dell sell machines with Ubuntu installed? After all, all of those people and Dell could use any distro they want. Yet, they don't, they choose Ubuntu despite the problems. Do these problems stop a few people adopting Linux? Probably. But does not Ubuntu also attract the multitudes?

I will read your next article on a computer running Ubuntu, but doubt that I will bother to comment.

ps

One thing we can agree on is that Sam Varghese's article was unwarranted.

It wasn't one half of one percent of Intel graphics chipset users. It was the vast majority. That is very well documented and it refers to all users of the affected versions of the Intel drivers, not just Ubuntu users. Stop trying to minimize something that was major with no facts to back up your assertions. Good enough for millions of users? If they didn't have Intel chipsets, sure. Failure for millions of users? Almost certainly.

If the fix required a new version mini-upgrade than that is precisely what should have been done. Actually, that version of Ubuntu should NEVER have included such severely broken code. Your "minor distros" included Fedora and Mandriva. I guess to you anyone who isn't Ubuntu is "minor".

If you want to setup an interview with Jono Bacon I'd be happy to interview him. Otherwise, no, I have no plans to rehash an issue that has long been solved. I already linked to Chris Smart's interview with Jono Bacon. Do you really think his answers will have changed?

Anyway, I completely disagree with your conclusions. I already pointed out that Ubuntu works with manufacturers to perfect releases on specific machines so your Dell example is a red herring. I even complimented Canonical for that in the previous article and mentioned it in this one.

Ubuntu doesn't attract multitudes. That's nonsense through and through. Preloaded machines attract the masses. The original EeePC offering with Xandros Presto, as flawed as Xandros is, did more than Ubuntu. Ditto the Acer Aspire One offering with Linpus. Those were the products that got Linux into mainstream stores for a year or more and are the base Dell and others, including Ubuntu, built on. As I've already noted, Ubuntu, when preloaded, is certainly better that Xandros or Linpus. It's the wider, downloadable general release that is problematic. Of course, I've said all of this before and you chose to ignore it then so I am sure you'll blithely ignore it now.

Ubuntu does attract some Windows geeks to Linux. If it fails for them and they equate Linux with Ubuntu they go on to be the people who proclaim how much "Linux sucks" to the rest of the world. That is the impact of Ubuntu as a standard bearer for Linux.

Oh well, Varghese is a punk. What else would you expect?

Visited O'Reilly's site to see if they had re-instated their entry-level bookshelf, but sadly not, still $23 a month, minimum.

Anyway, stumbled upon this "heated" Linux debate. Mr Varghese sounds like a crawler - he clearly isn't the type to attack a product made by someone with cash to burn. Mr Varghese should become a politician - then he can devote years of his life to pleasing the big name CEOs and Wall Street types who care more about money than human life.

I like Linux, but I agree, it has faults. The Linux community, however, should NOT emulate Microsoft's arrogance and deny there are problems.

I should have added that, far from Mr Varghese's belief that all Linux users have a unmitigated sense of entitlement - and think the world owes them a living? - I paid for my version of Linux. I'm happy to pay for Linux IF it means a commercially viable alternative to Windows can be developed.

Mr Varghese states that Mark Shuttleworth does NOT have to listen to anyone. That being the case, perhaps the Ubuntu name should be changed: to "Mark Shuttleworth's plaything". At least, then, the Linux community will know what kind of distro they are dealing with.

Mr Varghese has quite put me off Ubuntu.

I was very surprised by Mr Varghese's blog, full of wisdom, as he himself recommended it (in another post, it was how he self-qualified his work):
even during Christmas, intervals between his posts never were greater than three days, and he somtimes posted 2 times a day on IT-wire. I am very surprised he did not post anything there since Apr, 17 and moderation was left to his hierarchy, as if he were an autotroll;
But perhaps
a) he is exhausted,
b) nothing happened in the free world since Apr,17 (a mega freeze?)
c) he broke his precious little fingers?
d) there is no wisdom left...

News Topics

Recommended for You

Got a Question?