"And there was nothing about Linux in the accompanying agreement"
If you are going to try and undermine competitors in an anticompetitive fashion, probably best not to spell it out in black and white. Not that I'm suggesting this is what Microsoft did, but if they did, I'm sure the Microsoft entrance tests are good enough to filter out anyone stupid enough to write it down (Halloween docs?).
Personally I think Word Perfect sucked on Unix, and I assume the GNU/Linux version sucked as well and in the same fashion. They were competing with apps like Applixware (or whatever it became) at the top, and Star Office at the low end. So you could have gratis and better, or more expensive but superb whole Office suite, with WP squeezed in the middle. Bit of a no brainer, you'd have to have been a Word Perfect diehard to stick at it, and I don't think there were many left by then.
Corel Linux - don't think I ever saw it - marketing issue there probably.
But I don't think there has ever been much money in Desktop GNU/Linux other than for support, and I predict there won't be until it consists of more than about 1% of the boxes visiting my websites (even the local GLUG site is still running 55% Windows to 23% Linux (19% unknown) -- one day I'll have to look at what the Windows users are reading on the site - perhaps they are all bots).
Mod parent up -- oh he had 5 points -- still give it a try.
Please complete the following;
We need another enterprise file system;
- like we need another web browser. - like we need another Window manager. - like we need another Bourne shell derivative. - more than we need improved network filesystem support. - more than we need Hans Reiser to rip out the limitations of the VFS from the kernel. - because the other Enterprise level filesystems just don't support big enough filesystems/files. - because the other Enterprise level filesystems are too fast. - because backwards compatibility to ext2 is written in stone on tablets handed to McKusick on Mount Berkeley by someone with a burning beard. - because cludgy journaling addon are cool and we should strive to preserve cludges as long as possible. - other - please specify.
The GNOME menu is cunningly accompanied by the word "Applications".
As in "Applications" > "System Tools" > "File Browser". As opposed to say "Start" > "Accessories" > "Windows Explorer"
No one mention the "I have to click 'start' to shutdown" story;)
The guy didn't report particular problems using his GNU/Linux desktop, he reports problems importing/exporting Word documents (Microsoft keep the format secret to discourage competition). Handling DVDs (the DVD consortium want you to paid them so you can have a player for your DVD).
And some issues with hardware recognitions, and media formats. Again the media formats are largely a proprietary format issue, and the hardware recognition largely comes down to industry support. The predominant difference between installing Windows and GNU/Linux these days, is that usually someone else installs Windows for you, and ships you a reinstall disk.
Mostly the story screams to me "don't buy into proprietary data formats" because you'll be locked into paying the same supplier no matter how expensive their product, how slow the release schedule, or how poor the quality. Guess it is a bit late to tell people what most good IT managers had learnt by the 1970's.
Don't buy media with daft copy protections schemes, which are designed to rake more money out of the consumer (DVD regions anyone?), because they'll rake money out of you.
I was thinking of writing an article myself on why GNU/Linux hasn't found widespread adoption, but I don't think it is simply an ease of use thing. However the reputation for being "hard to use" may contribute.
And I certainly agree a completely free software GNU/Linux desktop has issues with the current plethora of Flash, and other rapidly changing formats, if you are happy to bung in proprietary components for Flash, Java and such like, which some distros do, and get it preinstalled, I think many more converts could be made.
It is a great pity, as the underlying technologies in many free software operating systems do make Microsoft look pretty mickey mouse by comparison.
Sorting an (a known issue -- I lept into the deep end) issue with a cutting edge version of GNU/Linux the other day, I uninstalled and reinstalled 1400 graphical applications, which required almost no manual interaction, no reboots (I said applications, not operating system changes), no accepting of licence agreements, or entering of license keys. I couldn't even conceive of anything close to this under Windows, without requiring a full format and reinstall and a lot of time, keys, and clicking.
Recently getting a Windows XP box back to the level of performance it should have on the box in question required 3 reboots for what shouldn't have gone wrong, and could have been fixed in GNU/Linux with one command. The underlying bug (a problem with how XP handles errors for IDE devices) is serious, basically unfixed as the Microsoft's "fix" just makes the issues less common, and presumably is slowing down an awful lot of PCs out there with less clueful Admins/owners.
Better yet I quickly established it was a software issue by booting with a LiveCD (Yes you guessed it GNU/Linux). What was really scary was the LiveCD could run 40 odd simultaneous multimedia apps on the hardware at the same time (from CD) smoothly, where as even when it was working correctly XP struggled to get passed one or two without getting a bad case of the "Max Headroom's".
But I'd have to concur that the free software desktop experience is still lagging slightly (when Windows works that is). What's more I don't expect that to change, until and unless it gets widespread adoption, at least in some parts of the world, as until that happens the Adobe's and Intels of this world will treat it as a second class citizen. Hopefully India, China or Latin America will be the place it happens, but I'm not that optimistic any more.
All the original story says was the account was closed to try and mitigate the attack. It doesn't say if Blue Security were involved in the decision to close the account or not.
I know from comments by TUCOWS that the domain wasn't registered with them, but presumably was hosted on their DNS servers (hence the attack).
Moving such services around at such a time would make sense from both ends, as it might restore the service for the client, as well as give TUCOWS a chance to get back to serving the rest of the domains it host.
I work for a TUCOWS reseller, but we were unaffected by the incident as we host our own DNS.
> With distro's like Ubuntu releasing every 6 months (yes, I know about Dapper's delay) with a relatively STABLE product with cutting edge features (Gnome 2.14), is it time for Debian to rethink their policy.
"relatively stable" is like "kind of pregnant", first Ubuntu release I put into production was replaced inside a month with Sarge, because Sarge worked and Ubuntu didn't.
Which is why the article is wrong, Etch will release when it is ready, which means when all the important bugs we know about are fixed (within a little margin of error), and not a day earlier.
What the article is really saying is that this release is going well, and on target, and I know Ubuntu's efforts have helped get some big bits of functionality into Etch quicker than they otherwise would have, but it takes time to iron out the bugs in big software projects.
I wonder what the burn rate is on Ubuntu these days?
I don't think so -- I think it is simply dearth of experience in the area.
Whilst ISPs have been using free software solutions for user management for years, and love it because they can easily integrate any old third party software without coughing up money. You could probably group all the people who deployed such in a small stadium.
Skill in big directory services type skill on Unix/Linux is pretty sparse on the ground. Probably a lot more people like me around who've done integration with relational database for authentication and authorisation, done NIS, done NIS+, but look at LDAP and Kerbero,s and go "wibble wibble argh". I do PAM now and then as a Linux admin, but never frequently enough to say I'm know it, it is set and forget for most small companies, and most specific server set ups.
Heck even integrating "samba" into ADS has a pretty big learning curve. On the upside the people who have done that understand what the Microsoft systems are doing underneath (often this is the reason they are die hard GNU/Linux bigots).
I think this is reflected in the story, the company which is doing the big service for lots of users, loves the "we can roll our own solution" approach, as they only have to do it once for 2,000,000 users, for a limited range of apps.
The others are trying to find the skill to do these kind of things. Although if someone stuck email on the desktops in 1999 they were pretty lame, I was deploying IMAP4 on my home network by then, to avoid keeping email in proprietary formats on the desktop.
On the other side of the coin, whilst Microsoft have put some nice GUIs on top of ADS, the number of people who really understand Microsoft ADS and the security model is surprisingly small, and clearly didn't include a lot of people who worked on the built-in services for Windows 2000 at Microsoft. So a case of "its fine when it is working".
I'm very surprised at the comment about having to spend time managing GNU/Linux systems. Our experience is that it is the Windows boxes that "suddenly" do something odd, because of the lack of transparency, and poor logging (what you mean it can log filesystem corruption without mentioning which filesystem got corrupted ?). Similarly surprised at comments on server consolidation, my experience has always been that Windows almost invariably ends up being deployed on more boxes, either for performance reasons, or to separate services because of the frequent needs for reboots when applications are updated.
I suspect a lot of the "benefits" here would have accrued, be throwing out the old, and bringing in the new, whatever the new was.
> If you have a Win98 machine and it is crashing, you have crap third party apps, or a crap driver.
Windows 98 (and Windows 95) did have some serious issues unrelated to third party applications. For example the dial-up software shipped with it was quite capable of corrupting itself in hideous ways (writing all over its own config files was one favourite trick). Don't anyone mention the famed 49.7 days bug, although technically that was a "crap driver" I suppose, it wasn't a third party driver.
If you have an OS that doesn't protect itself from rogue applications, it doesn't matter who wrote the application, merely what they wrote. I had similar issues with Netware 3, and Novell NLMs that crashed servers, because there was no proper process protection.
Just as I stopped running the Novell NLMs that caused our servers to crash, people learnt to live with Windows 95/98, because in other regards it met their needs (I know I ran Windows 95 for 11 years because it was good enough). Alas a whole generation now thinks it is relatively normal for "computers" to be unreliable, need rebooting for cosmetic changes etc.
My desktop box just had the entire graphical desktop, and every graphical application uninstalled, and reinstalled (long story), no reboots required, no OS hangs. Throughout this rather irritating (and unusual) episode it continued to serve email, files, and other minor services to other machines on the local network without even a hiccup, as well as utilising the Internet to install the relevant versions of the software that were being sorted out. However I appreciate this experience is likely to be very different from that of the majority of computer users, it illustrates the power and potential of proper layering and isolation of software components.
Recently was surprised to find one of my mates was trying SME server inspite of a limited Linux background. But it is based on ancient Redhat version, and it shows, and I can't recommend it because it is so dated.
There is definitely room for "based on Debian" server platforms. Xandros have things like email, and web servers pretty much for free, all they need to sort is the user management, which they will need to sort out anyway if they hope to move out of small site deployments on the desktop into larger deployments. Unless you build a good peer to peer style Office network system, you really have to provide both a server and client solution if you hope to do "desktops" these days.
I think the market for small Office server appliance is still under served as well.
> how much outside of the basic gnu tool chain and the gcc is gnu made today? there is kde and gnome
GNOME is part of the GNU project -- if you are going to belittle the GNU project contribution, at least have the decency to know what you are talking about.
RMSs personal contribution include a hell of a lot of code, and even more inspiring other people.
Personally I think we should call the GNU project OS "Debian" for short, but I'm not sure I'm carrying many people with me on that one.
Simon (sometime GNU maintainer whose own contributions would be a far easier target for mocking)
I seem to recall someone analysing the code, saying it stepped through and stat'd every font file on start-up or some similar totally unnecessary programming sin (since GNOME has already done all that sort of thing). As a result it was slow to start-up, not slow in use, although that would depend hugely on things like number of fonts installed, and system performance. Especially the first start-up after a reboot (whatever they are).
I suspect one of those matters of programming honour, once someone pointed out how hideously inefficient the process was, even if from a practical perspective it mattered very little.
time for a in 1 2 3 4 5 6 7 8 9 0 > do > gnome-terminal -e "exit" > done
real 0m6.898s user 0m4.032s sys 0m0.396s srw@derek:~$ time for a in 1 2 3 4 5 6 7 8 9 0; do xterm -e "exit"; done
"I still don't understand what makes apt-get years ahead of something like urpmi. Both seem to just work, and do about the same sort of thing. I don't have problems with either."
I don't think "apt-get" was ever that revolutionary in a technical sense. Coming from an HP-UX background, I was using Software Distributor, that did a lot of the things that "apt" did years ago, including clearing out superseeded packages from repositories, and such like. Okay it wasn't as "web ready", but then the web wasn't so important then.
The thing that distinguished Debian is both the quality of the packages (in terms of how well packaged, not the software contained), the variety of packages (in official archives with set standards of quality and support), and that they have long been used with reliance on the dependencies, and their automatic resolution. So when you type "apt-get install libapache2-mod-perl2" you do get the right set of packages underneath to make it "just work".
Sure the tools to do this in the RPM world have been catching up, but I doubt if you take a vanilla system minimal install with most of these distros, and type a selection of similar commands to the above, that you'll get them all to "just work" as slickly as Debian Sarge will. But that is nothing to do with the tool itself, just the data it has to work with. Although some of the RPM tools can be painfully slow (don't mention yum).
I think it is more fundamental - Office suites just aren't good for what people use them for. Some of the presentation tools aren't that bad, but I think the whole presentation thing has evolved to fit Powerpoint, so perhaps I should judge them by how well they communicate (export to PDF, export to Web, show annotations).
I've only worked at a handful of companies that had meaningful integration between their Office suite, and say the customer database, and quite a few of those were running Office Automaton (the younger folk can look it up elsewhere).
Integration, and fitness for purpose are key, and I think many places would be far better off with decent letter writing tools.
My main use of the Office suite on my PC is letter writing, and the Wizards does a vague job of filling in the address information, date, time. The letter templates are rubbish, and very limited, and the wizard doesn't really address storing the letters (so I manually save them in the letters folder, "YYYYMMDDrecipient.ext").
My main other use is structured documents for reports/documentation, and none of the Office suites do this terribly well, hence all the Latex/Web stuff earlier.
If we just admitted this, broke the common tasks down further, and had 10 simple apps, I'd be happier and I'm sure others would too.
However I would note one of our local schools blessed with an IT chap, who gets paid a pittance, and is worth, and could earn far more elsewhere. If I ever have another IT company at the point of employing someone, poaching same is not beyond the bounds of possibility.
You think IBM made big kernel contributions when it mattered?
The Linux kernel started from individuals it probably would never have happened the way it did without AT&T, but I don't think there was any big corporate contributions till it became big on merit. Sure people started companies from Linux, and free software, I think that was probably as common as big corporate sponsorship.
I don't want to knock IBMs contributions, but I think they came quite late to the table, okay they brought a lot to the table when they came, but Linux was already successful by almost any criteria.
Similarly other open source or free software efforts without major corporates would include Apache, Sendmail, and a whole host of similar applications that made the Internet possible. More recently qmail was largely down to one individual. Seems to me big Academia can contribute to such projects far more easily than big business.
I don't see the hand of big business in MySQL, or a lot of other places. It is hard to say for sure unless you know how a project was funded. For example I don't know if Postfix happened because of, or inspite of, IBM. I suspect a lot of projects were done by people employed by big companies inspite of the big companies.
And whilst I think capitalism requires capitalists, I'm not sure the current big multinationals are either healthy, or necessary.
Debian is huge, but last time the SPI accounts were waved at me they had less money in the bank than I did (a situation that I believe had since changed -- mine for the worse [che sera sera]). Sure these projects aren't a free lunch, but you can do big things if a lot of people contribute their little piece. Debian has download servers all over the world, few if any belong to any identifiable "Debian" organisation.
Many of these big companies were built taking a small amount from many individuals, these projects prove that for certain things (mainly software) you can do similar things by volunteer effort.
"What is it going to take for these studies to finally start comparing apples to apples in regards as to what really is part of the OS and what is required for it to run?".
When Windows Update patches all the third party applications installed on a box, we can then compare the five minutes it takes the GNU/Linux admins to handle security patches to how long it takes in Windows. Of course by then it'll only be taking the Windows admins 5 minutes a week.
Simplistic answers are out.
I can install Windows 2003, and get a pretty neat config for Office file sharing. To do the same thing in Debian would take me a lot longer, well if I had to reproduce the "previous version" type functionality.
However once I'd done it in Debian I could reproduce it faster, and it wouldn't require new client software on every PC in the network, and the versioning would be proper "copy on write", using less disk space, and allowing finer control on which copies are retained. And I could then give it away as a "boot and go" CD to my clients. Similarly at the other end of the lifespan of such a system, when something goes wrong I'm more confident Debian will give me meaningful error messages.
Which is better there? I'd say the free software for moral reasons - but that isn't a simple argument either. I can see why a lot of small businesses would go for Windows with that requirement, even if they end up regretting it in the long run.
Curiously as Windows becomes more viable on the server, it is almost impossible for the small business to secure their Windows desktops effectively. I can imagine comapnies migrating to Windows as a directory service provider with thin client GNU/Linux desktops.
Okay - assuming equal pre-effort in equipment choice, and equal exposure to both platforms - I'm betting any modern GNU/Linux distro is easier to install and easier to install software on. Well you did ask me to guess.
Without OEMed hardware both can be a major pain to install, but Windows is the only one I ever had to custom build my own install disk for, and that required third party proprietary tools for writing the CD image as no one had documented how to do without them.
Great review, interesting book, I'll consider buying a copy.
But the name "Patch Management" sorry that really grates on me. Almost universally GNU/Linux systems have abandoned patches, and perform upgrades to whole components at a logical level. Its the best way to do it found so far, but I don't think of those as "patches".
"Sure, Apple needs to patch up their XML so it's more correct... "
The well formed nature of XML is a yes/no decision. So it can't be more correct, it can only be right or wrong. Slightly pregnant?
If your parser of the XML is correctly standard conforming it will not work with these feeds (assuming the article is correct), if yours does file a bug report. Just because XML is well formed doesn't make it useful, meaningful or even complying to RSS standards, but if it isn't well formed it is just corrupt data.
Earlier reference to HTML miss the point, the XML standards are designed inherently to be strict about the structure of documents. It is this kind of strictness that will allow us to produce small, fast, robust, interoperable and secure software for handling such data.
"When you're adding technology such as photocasting into an existing product, and such functionality it isn't necessary covered by the standard, you may have no choice but to create a standards-incompatible product."
The X in XML is from eXtensible....
Sure there maybe technologies that don't fit XML well, but if a document declares itself to be XML and isn't valid, then someone goofed.
A bit like saying...
'This sentence is all in British English, so warum kannst du nicht es verstanden, merci ?'
"You won't hear a civil engineer bitch about how annoying earthquakes are"
You would if the earthquakes were being created by the civil engineers in the next Office. Especially if it was because the civil engineers in the next Office were incompetent.
This is a very elementary mistake to have made. The document is in a standard format, heck the first line of the format explains that in pretty readable text, and the programmers obviously have issues with understanding that, or possibly the marketing people shipped a prototype. Either way reflects very badly on Apples professionalism.
Is Debian that much better whe it comes for day to day operations?
Stevey lacks enough karma so I'll reply as well;)
Yes. Assuming by day to day operations you mean installing and configuring software, which is what system admins tend to do.
Oh you wanted a longer response? Debian is very easy to manage because the packages fit and work well together. If you want "Postfix" with a "postgres" backend for authentication, you type "apt-get install postfix-pgsql" (or use a GUI if you must), the packaging tools figure out the dependencies. The Postgres server isn't a dependency, as the database might be on another server, but replacing your current incompatible MTA (usually Exim) would be an action performed, if necessary, after it asks if you are sure about the derived changes it wants to make.
Most other distros would expect you to remove the existing MTA yourself, and probably don't have a prebuilt package for this (admittedly fairly unusual) task, and you'd probably have to build Postfix, or a package for Postfix, from source, with some arcane './configure --with-pgsql --with-ssl --with-losing-will-to-use-free-software --with-wondering-why-we-didn't-buy-exchange' options.
Superficially this looks like a good packaging system, but the submerged iceberg mentioned in the article is 16,000 packages almost universally high quality packages that make this possible. For every MTA to replace the old MTA the package must declare it is an MTA etc etc.
Generally these packages are standard free software packages, but the Debian developers will make little changes to make them consistent and flexible. For example Postgres comes with a slightly tightened security set up as the source built one is rather insecure for most hosting environment. I dare say a lot of servers are just those little bit better configured for those little touches to the postgres security config. That consistency includes something in/usr/share/doc/packagename that at worst will include a URL to the documentation, and a file explaining any Debian customisations or perculiarities.
Even where Debian does push you out to compiling stuff from source (and you can build some pretty crazy and complex systems without going to source), the packaging system makes it easy to pull in the tools to do it. Need a new kernel, a package exists to help you make kernel packages easily from source, and Debian also provides kernel source packages with little customisations already applied.
I've no doubt Redhat has improved immensely in these areas since the versions we have, but the selling point with Redhat is the commercial support, and standard platform nature for certain applications. For certain enterprise applications, like Oracle, running outside the vendor supported platform is a no-no. So you have to buy a standard platform. Almost all big commercial applications support Redhat, most will run on Debian, but if your paying big money for applications the bosses expect proper vendor support and so it is Redhat. The ISP business is more flexible generally in these areas, and Debian is huge in this area.
My main gripe with the article, is I'd happily recommend Debian for the beginner. I wouldn't recommend they install it themselves, or configure it themselves, but if a relative wants a computer to surf and email, and print, and you want to set something up and forget about it, and yet still be able to install new software remotely with ease and speed.
The really funny bit is the article talks about expensive source code analysis tools that commercial companies often use...
Well I can assure you from many years walking through the door of software companies, that proper software checks are rarely run in private industry. If you are lucky the programmer will deal with the compiler warnings for a quiet life.
Still it is good someone is looking, wonder what David Wheeler could have done with the money?
Slashdot Mirror
"And there was nothing about Linux in the accompanying agreement"
If you are going to try and undermine competitors in an anticompetitive fashion, probably best not to spell it out in black and white. Not that I'm suggesting this is what Microsoft did, but if they did, I'm sure the Microsoft entrance tests are good enough to filter out anyone stupid enough to write it down (Halloween docs?).
Personally I think Word Perfect sucked on Unix, and I assume the GNU/Linux version sucked as well and in the same fashion. They were competing with apps like Applixware (or whatever it became) at the top, and Star Office at the low end. So you could have gratis and better, or more expensive but superb whole Office suite, with WP squeezed in the middle. Bit of a no brainer, you'd have to have been a Word Perfect diehard to stick at it, and I don't think there were many left by then.
Corel Linux - don't think I ever saw it - marketing issue there probably.
But I don't think there has ever been much money in Desktop GNU/Linux other than for support, and I predict there won't be until it consists of more than about 1% of the boxes visiting my websites (even the local GLUG site is still running 55% Windows to 23% Linux (19% unknown) -- one day I'll have to look at what the Windows users are reading on the site - perhaps they are all bots).
Mod parent up -- oh he had 5 points -- still give it a try.
Please complete the following;
We need another enterprise file system;
- like we need another web browser.
- like we need another Window manager.
- like we need another Bourne shell derivative.
- more than we need improved network filesystem support.
- more than we need Hans Reiser to rip out the limitations of the VFS from the kernel.
- because the other Enterprise level filesystems just don't support big enough filesystems/files.
- because the other Enterprise level filesystems are too fast.
- because backwards compatibility to ext2 is written in stone on tablets handed to McKusick on Mount Berkeley by someone with a burning beard.
- because cludgy journaling addon are cool and we should strive to preserve cludges as long as possible.
- other - please specify.
The GNOME menu is cunningly accompanied by the word "Applications".
;)
As in "Applications" > "System Tools" > "File Browser".
As opposed to say "Start" > "Accessories" > "Windows Explorer"
No one mention the "I have to click 'start' to shutdown" story
The guy didn't report particular problems using his GNU/Linux desktop, he reports problems importing/exporting Word documents (Microsoft keep the format secret to discourage competition). Handling DVDs (the DVD consortium want you to paid them so you can have a player for your DVD).
And some issues with hardware recognitions, and media formats. Again the media formats are largely a proprietary format issue, and the hardware recognition largely comes down to industry support. The predominant difference between installing Windows and GNU/Linux these days, is that usually someone else installs Windows for you, and ships you a reinstall disk.
Mostly the story screams to me "don't buy into proprietary data formats" because you'll be locked into paying the same supplier no matter how expensive their product, how slow the release schedule, or how poor the quality. Guess it is a bit late to tell people what most good IT managers had learnt by the 1970's.
Don't buy media with daft copy protections schemes, which are designed to rake more money out of the consumer (DVD regions anyone?), because they'll rake money out of you.
I was thinking of writing an article myself on why GNU/Linux hasn't found widespread adoption, but I don't think it is simply an ease of use thing. However the reputation for being "hard to use" may contribute.
And I certainly agree a completely free software GNU/Linux desktop has issues with the current plethora of Flash, and other rapidly changing formats, if you are happy to bung in proprietary components for Flash, Java and such like, which some distros do, and get it preinstalled, I think many more converts could be made.
It is a great pity, as the underlying technologies in many free software operating systems do make Microsoft look pretty mickey mouse by comparison.
Sorting an (a known issue -- I lept into the deep end) issue with a cutting edge version of GNU/Linux the other day, I uninstalled and reinstalled 1400 graphical applications, which required almost no manual interaction, no reboots (I said applications, not operating system changes), no accepting of licence agreements, or entering of license keys. I couldn't even conceive of anything close to this under Windows, without requiring a full format and reinstall and a lot of time, keys, and clicking.
Recently getting a Windows XP box back to the level of performance it should have on the box in question required 3 reboots for what shouldn't have gone wrong, and could have been fixed in GNU/Linux with one command. The underlying bug (a problem with how XP handles errors for IDE devices) is serious, basically unfixed as the Microsoft's "fix" just makes the issues less common, and presumably is slowing down an awful lot of PCs out there with less clueful Admins/owners.
Better yet I quickly established it was a software issue by booting with a LiveCD (Yes you guessed it GNU/Linux). What was really scary was the LiveCD could run 40 odd simultaneous multimedia apps on the hardware at the same time (from CD) smoothly, where as even when it was working correctly XP struggled to get passed one or two without getting a bad case of the "Max Headroom's".
But I'd have to concur that the free software desktop experience is still lagging slightly (when Windows works that is). What's more I don't expect that to change, until and unless it gets widespread adoption, at least in some parts of the world, as until that happens the Adobe's and Intels of this world will treat it as a second class citizen. Hopefully India, China or Latin America will be the place it happens, but I'm not that optimistic any more.
All the original story says was the account was closed to try and mitigate the attack. It doesn't say if Blue Security were involved in the decision to close the account or not.
I know from comments by TUCOWS that the domain wasn't registered with them, but presumably was hosted on their DNS servers (hence the attack).
Moving such services around at such a time would make sense from both ends, as it might restore the service for the client, as well as give TUCOWS a chance to get back to serving the rest of the domains it host.
I work for a TUCOWS reseller, but we were unaffected by the incident as we host our own DNS.
> With distro's like Ubuntu releasing every 6 months (yes, I know about Dapper's delay) with a relatively STABLE product with cutting edge features (Gnome 2.14), is it time for Debian to rethink their policy.
"relatively stable" is like "kind of pregnant", first Ubuntu release I put into production was replaced inside a month with Sarge, because Sarge worked and Ubuntu didn't.
Which is why the article is wrong, Etch will release when it is ready, which means when all the important bugs we know about are fixed (within a little margin of error), and not a day earlier.
What the article is really saying is that this release is going well, and on target, and I know Ubuntu's efforts have helped get some big bits of functionality into Etch quicker than they otherwise would have, but it takes time to iron out the bugs in big software projects.
I wonder what the burn rate is on Ubuntu these days?
I don't think so -- I think it is simply dearth of experience in the area.
Whilst ISPs have been using free software solutions for user management for years, and love it because they can easily integrate any old third party software without coughing up money. You could probably group all the people who deployed such in a small stadium.
Skill in big directory services type skill on Unix/Linux is pretty sparse on the ground. Probably a lot more people like me around who've done integration with relational database for authentication and authorisation, done NIS, done NIS+, but look at LDAP and Kerbero,s and go "wibble wibble argh". I do PAM now and then as a Linux admin, but never frequently enough to say I'm know it, it is set and forget for most small companies, and most specific server set ups.
Heck even integrating "samba" into ADS has a pretty big learning curve. On the upside the people who have done that understand what the Microsoft systems are doing underneath (often this is the reason they are die hard GNU/Linux bigots).
I think this is reflected in the story, the company which is doing the big service for lots of users, loves the "we can roll our own solution" approach, as they only have to do it once for 2,000,000 users, for a limited range of apps.
The others are trying to find the skill to do these kind of things. Although if someone stuck email on the desktops in 1999 they were pretty lame, I was deploying IMAP4 on my home network by then, to avoid keeping email in proprietary formats on the desktop.
On the other side of the coin, whilst Microsoft have put some nice GUIs on top of ADS, the number of people who really understand Microsoft ADS and the security model is surprisingly small, and clearly didn't include a lot of people who worked on the built-in services for Windows 2000 at Microsoft. So a case of "its fine when it is working".
I'm very surprised at the comment about having to spend time managing GNU/Linux systems. Our experience is that it is the Windows boxes that "suddenly" do something odd, because of the lack of transparency, and poor logging (what you mean it can log filesystem corruption without mentioning which filesystem got corrupted ?). Similarly surprised at comments on server consolidation, my experience has always been that Windows almost invariably ends up being deployed on more boxes, either for performance reasons, or to separate services because of the frequent needs for reboots when applications are updated.
I suspect a lot of the "benefits" here would have accrued, be throwing out the old, and bringing in the new, whatever the new was.
> If you have a Win98 machine and it is crashing, you have crap third party apps, or a crap driver.
Windows 98 (and Windows 95) did have some serious issues unrelated to third party applications. For example the dial-up software shipped with it was quite capable of corrupting itself in hideous ways (writing all over its own config files was one favourite trick). Don't anyone mention the famed 49.7 days bug, although technically that was a "crap driver" I suppose, it wasn't a third party driver.
If you have an OS that doesn't protect itself from rogue applications, it doesn't matter who wrote the application, merely what they wrote. I had similar issues with Netware 3, and Novell NLMs that crashed servers, because there was no proper process protection.
Just as I stopped running the Novell NLMs that caused our servers to crash, people learnt to live with Windows 95/98, because in other regards it met their needs (I know I ran Windows 95 for 11 years because it was good enough). Alas a whole generation now thinks it is relatively normal for "computers" to be unreliable, need rebooting for cosmetic changes etc.
My desktop box just had the entire graphical desktop, and every graphical application uninstalled, and reinstalled (long story), no reboots required, no OS hangs. Throughout this rather irritating (and unusual) episode it continued to serve email, files, and other minor services to other machines on the local network without even a hiccup, as well as utilising the Internet to install the relevant versions of the software that were being sorted out. However I appreciate this experience is likely to be very different from that of the majority of computer users, it illustrates the power and potential of proper layering and isolation of software components.
Agreed whole heartedly.
Recently was surprised to find one of my mates was trying SME server inspite of a limited Linux background. But it is based on ancient Redhat version, and it shows, and I can't recommend it because it is so dated.
There is definitely room for "based on Debian" server platforms. Xandros have things like email, and web servers pretty much for free, all they need to sort is the user management, which they will need to sort out anyway if they hope to move out of small site deployments on the desktop into larger deployments. Unless you build a good peer to peer style Office network system, you really have to provide both a server and client solution if you hope to do "desktops" these days.
I think the market for small Office server appliance is still under served as well.
> how much outside of the basic gnu tool chain and the gcc is gnu made today? there is kde and gnome
GNOME is part of the GNU project -- if you are going to belittle the GNU project contribution, at least have the decency to know what you are talking about.
RMSs personal contribution include a hell of a lot of code, and even more inspiring other people.
Personally I think we should call the GNU project OS "Debian" for short, but I'm not sure I'm carrying many people with me on that one.
Simon (sometime GNU maintainer whose own contributions would be a far easier target for mocking)
I seem to recall someone analysing the code, saying it stepped through and stat'd every font file on start-up or some similar totally unnecessary programming sin (since GNOME has already done all that sort of thing). As a result it was slow to start-up, not slow in use, although that would depend hugely on things like number of fonts installed, and system performance. Especially the first start-up after a reboot (whatever they are).
I suspect one of those matters of programming honour, once someone pointed out how hideously inefficient the process was, even if from a practical perspective it mattered very little.
time for a in 1 2 3 4 5 6 7 8 9 0
> do
> gnome-terminal -e "exit"
> done
real 0m6.898s
user 0m4.032s
sys 0m0.396s
srw@derek:~$ time for a in 1 2 3 4 5 6 7 8 9 0; do xterm -e "exit"; done
real 0m2.003s
user 0m0.280s
sys 0m0.068s
Oops user CPU increased by 20 fold over xterm.
"I still don't understand what makes apt-get years ahead of something like urpmi. Both seem to just work, and do about the same sort of thing. I don't have problems with either."
I don't think "apt-get" was ever that revolutionary in a technical sense. Coming from an HP-UX background, I was using Software Distributor, that did a lot of the things that "apt" did years ago, including clearing out superseeded packages from repositories, and such like. Okay it wasn't as "web ready", but then the web wasn't so important then.
The thing that distinguished Debian is both the quality of the packages (in terms of how well packaged, not the software contained), the variety of packages (in official archives with set standards of quality and support), and that they have long been used with reliance on the dependencies, and their automatic resolution. So when you type "apt-get install libapache2-mod-perl2" you do get the right set of packages underneath to make it "just work".
Sure the tools to do this in the RPM world have been catching up, but I doubt if you take a vanilla system minimal install with most of these distros, and type a selection of similar commands to the above, that you'll get them all to "just work" as slickly as Debian Sarge will. But that is nothing to do with the tool itself, just the data it has to work with. Although some of the RPM tools can be painfully slow (don't mention yum).
I think it is more fundamental - Office suites just aren't good for what people use them for. Some of the presentation tools aren't that bad, but I think the whole presentation thing has evolved to fit Powerpoint, so perhaps I should judge them by how well they communicate (export to PDF, export to Web, show annotations).
I've only worked at a handful of companies that had meaningful integration between their Office suite, and say the customer database, and quite a few of those were running Office Automaton (the younger folk can look it up elsewhere).
Integration, and fitness for purpose are key, and I think many places would be far better off with decent letter writing tools.
My main use of the Office suite on my PC is letter writing, and the Wizards does a vague job of filling in the address information, date, time. The letter templates are rubbish, and very limited, and the wizard doesn't really address storing the letters (so I manually save them in the letters folder, "YYYYMMDDrecipient.ext").
My main other use is structured documents for reports/documentation, and none of the Office suites do this terribly well, hence all the Latex/Web stuff earlier.
If we just admitted this, broke the common tasks down further, and had 10 simple apps, I'd be happier and I'm sure others would too.
However I would note one of our local schools blessed with an IT chap, who gets paid a pittance, and is worth, and could earn far more elsewhere. If I ever have another IT company at the point of employing someone, poaching same is not beyond the bounds of possibility.
You think Debian is small?
You think IBM made big kernel contributions when it mattered?
The Linux kernel started from individuals it probably would never have happened the way it did without AT&T, but I don't think there was any big corporate contributions till it became big on merit. Sure people started companies from Linux, and free software, I think that was probably as common as big corporate sponsorship.
I don't want to knock IBMs contributions, but I think they came quite late to the table, okay they brought a lot to the table when they came, but Linux was already successful by almost any criteria.
Similarly other open source or free software efforts without major corporates would include Apache, Sendmail, and a whole host of similar applications that made the Internet possible. More recently qmail was largely down to one individual. Seems to me big Academia can contribute to such projects far more easily than big business.
I don't see the hand of big business in MySQL, or a lot of other places. It is hard to say for sure unless you know how a project was funded. For example I don't know if Postfix happened because of, or inspite of, IBM. I suspect a lot of projects were done by people employed by big companies inspite of the big companies.
And whilst I think capitalism requires capitalists, I'm not sure the current big multinationals are either healthy, or necessary.
Debian is huge, but last time the SPI accounts were waved at me they had less money in the bank than I did (a situation that I believe had since changed -- mine for the worse [che sera sera]). Sure these projects aren't a free lunch, but you can do big things if a lot of people contribute their little piece. Debian has download servers all over the world, few if any belong to any identifiable "Debian" organisation.
Many of these big companies were built taking a small amount from many individuals, these projects prove that for certain things (mainly software) you can do similar things by volunteer effort.
"which means he lost a _lot_ of money."
I suspect he'd mention that to the new employer when discussing renumeration, if it mattered.
"What is it going to take for these studies to finally start comparing apples to apples in regards as to what really is part of the OS and what is required for it to run?".
When Windows Update patches all the third party applications installed on a box, we can then compare the five minutes it takes the GNU/Linux admins to handle security patches to how long it takes in Windows. Of course by then it'll only be taking the Windows admins 5 minutes a week.
Simplistic answers are out.
I can install Windows 2003, and get a pretty neat config for Office file sharing. To do the same thing in Debian would take me a lot longer, well if I had to reproduce the "previous version" type functionality.
However once I'd done it in Debian I could reproduce it faster, and it wouldn't require new client software on every PC in the network, and the versioning would be proper "copy on write", using less disk space, and allowing finer control on which copies are retained. And I could then give it away as a "boot and go" CD to my clients. Similarly at the other end of the lifespan of such a system, when something goes wrong I'm more confident Debian will give me meaningful error messages.
Which is better there? I'd say the free software for moral reasons - but that isn't a simple argument either. I can see why a lot of small businesses would go for Windows with that requirement, even if they end up regretting it in the long run.
Curiously as Windows becomes more viable on the server, it is almost impossible for the small business to secure their Windows desktops effectively. I can imagine comapnies migrating to Windows as a directory service provider with thin client GNU/Linux desktops.
Okay - assuming equal pre-effort in equipment choice, and equal exposure to both platforms - I'm betting any modern GNU/Linux distro is easier to install and easier to install software on. Well you did ask me to guess.
Without OEMed hardware both can be a major pain to install, but Windows is the only one I ever had to custom build my own install disk for, and that required third party proprietary tools for writing the CD image as no one had documented how to do without them.
Great review, interesting book, I'll consider buying a copy.
But the name "Patch Management" sorry that really grates on me. Almost universally GNU/Linux systems have abandoned patches, and perform upgrades to whole components at a logical level. Its the best way to do it found so far, but I don't think of those as "patches".
Or is that just me?
"Sure, Apple needs to patch up their XML so it's more correct... "
The well formed nature of XML is a yes/no decision. So it can't be more correct, it can only be right or wrong. Slightly pregnant?
If your parser of the XML is correctly standard conforming it will not work with these feeds (assuming the article is correct), if yours does file a bug report. Just because XML is well formed doesn't make it useful, meaningful or even complying to RSS standards, but if it isn't well formed it is just corrupt data.
Earlier reference to HTML miss the point, the XML standards are designed inherently to be strict about the structure of documents. It is this kind of strictness that will allow us to produce small, fast, robust, interoperable and secure software for handling such data.
"When you're adding technology such as photocasting into an existing product, and such functionality it isn't necessary covered by the standard, you may have no choice but to create a standards-incompatible product."
The X in XML is from eXtensible....
Sure there maybe technologies that don't fit XML well, but if a document declares itself to be XML and isn't valid, then someone goofed.
A bit like saying...
'This sentence is all in British English, so warum kannst du nicht es verstanden, merci ?'
"You won't hear a civil engineer bitch about how annoying earthquakes are"
You would if the earthquakes were being created by the civil engineers in the next Office. Especially if it was because the civil engineers in the next Office were incompetent.
Go read what XML was designed for.
I think you are missing the point.
This is a very elementary mistake to have made. The document is in a standard format, heck the first line of the format explains that in pretty readable text, and the programmers obviously have issues with understanding that, or possibly the marketing people shipped a prototype. Either way reflects very badly on Apples professionalism.
?
Stevey lacks enough karma so I'll reply as well ;)
Yes. Assuming by day to day operations you mean installing and configuring software, which is what system admins tend to do.
Oh you wanted a longer response? Debian is very easy to manage because the packages fit and work well together. If you want "Postfix" with a "postgres" backend for authentication, you type "apt-get install postfix-pgsql" (or use a GUI if you must), the packaging tools figure out the dependencies. The Postgres server isn't a dependency, as the database might be on another server, but replacing your current incompatible MTA (usually Exim) would be an action performed, if necessary, after it asks if you are sure about the derived changes it wants to make.
Most other distros would expect you to remove the existing MTA yourself, and probably don't have a prebuilt package for this (admittedly fairly unusual) task, and you'd probably have to build Postfix, or a package for Postfix, from source, with some arcane './configure --with-pgsql --with-ssl --with-losing-will-to-use-free-software --with-wondering-why-we-didn't-buy-exchange' options.
Superficially this looks like a good packaging system, but the submerged iceberg mentioned in the article is 16,000 packages almost universally high quality packages that make this possible. For every MTA to replace the old MTA the package must declare it is an MTA etc etc.
Generally these packages are standard free software packages, but the Debian developers will make little changes to make them consistent and flexible. For example Postgres comes with a slightly tightened security set up as the source built one is rather insecure for most hosting environment. I dare say a lot of servers are just those little bit better configured for those little touches to the postgres security config. That consistency includes something in /usr/share/doc/packagename that at worst will include a URL to the documentation, and a file explaining any Debian customisations or perculiarities.
Even where Debian does push you out to compiling stuff from source (and you can build some pretty crazy and complex systems without going to source), the packaging system makes it easy to pull in the tools to do it. Need a new kernel, a package exists to help you make kernel packages easily from source, and Debian also provides kernel source packages with little customisations already applied.
I've no doubt Redhat has improved immensely in these areas since the versions we have, but the selling point with Redhat is the commercial support, and standard platform nature for certain applications. For certain enterprise applications, like Oracle, running outside the vendor supported platform is a no-no. So you have to buy a standard platform. Almost all big commercial applications support Redhat, most will run on Debian, but if your paying big money for applications the bosses expect proper vendor support and so it is Redhat. The ISP business is more flexible generally in these areas, and Debian is huge in this area.
My main gripe with the article, is I'd happily recommend Debian for the beginner. I wouldn't recommend they install it themselves, or configure it themselves, but if a relative wants a computer to surf and email, and print, and you want to set something up and forget about it, and yet still be able to install new software remotely with ease and speed.
The really funny bit is the article talks about expensive source code analysis tools that commercial companies often use...
Well I can assure you from many years walking through the door of software companies, that proper software checks are rarely run in private industry. If you are lucky the programmer will deal with the compiler warnings for a quiet life.
Still it is good someone is looking, wonder what David Wheeler could have done with the money?