That misses the point. A very large number of Linux users are in the position where they are consulted by others about what devices to buy. The availability of devices which will work with Linux increases a company's exposure to all kinds of user, not just Linux users.
Companies worried about IP issues should ask themselves if they are in the hardware business or the software business. If their objective is to sell more gizzmos, then opening the API to developers is an excellent way to sell more product.
If a company is concerned about the number of questions they'll be asked by the developers, then (a) they don't know the software business, and (b) they should take a long, hard look at the quality of their documentation.
The biggest problem is that many companies are already making so much from selling their gizzmos to Windows users not to need to sell them to Mac or Linux users as well, even though it takes no significant effort to do so. The extra profit, even at virtually 100% per unit) simply isn't attractive.
Just confirms my {subliminal:emacs} unshakeable {subliminal:emacs} prejudice {subliminal:emacs} that vi had its heyday some decades ago.
The time for dual-mode editors (where you have to press something before you can begin to type, and then press something else when you stop typing) is long since gone, thank goddess.
It's instructive to observe the panic-ridden frenzy with which Microsoft have approached the business of using XML as a file format. The marketing influence is all too plain to see, with the result that they feel an inner compulsion to preserve the appearance of the document at all costs, sacrificing all logic and common-sense to do it.
OOo did the same, but with greater elegance and less haste because they were ahead of the field. Corel screwed it up with WordPerfect by keeping their stylesheet format proprietary so that transfer between WP document code and XML was made as hard as possible (a Class A blunder, given that their XML editor is actually quite good). AbiWord makes a good job of saving DocBook XML, but it's not trying to pretend it's reimportable; it screws up LaTeX formidably, though, by trying to pretend that it absolutely has to preserve line-length and font-size, which is evidence of the same neurotic attitude as Microsoft.
The problem in all cases is not that the assorted authors and coders don't understand XML (although some of them clearly failed that test too), but that they don't understand documents. This is particularly true at Microsoft, where leaders such as Jean Paoli have been proselytizing XML for years. They still think a document is a jumble of letters; they have no idea of structure, and the DOM is simply laughable as a non-model of a document. Microsoft's particular problem with XML is that they came to it too late, and viewed it as a way of storing data, not text...indeed to this day many XML users, trained with Microsoft blinkers on, are unaware that XML can be used for normal text documents.
With this level of ignorance surrounding Microsoft, it's hardly unexpected that they should blunder so badly.
What's the big deal? Uncle Sam has been doing this for over a decade, sifting the applications for review using one of the major statistics packages, who developed special text-reporting tools for doing so.
I'd say over 50% of the forecasts will turn out -- but only in the USA. Some of them are already USA-specific anyway (eg TiVo) and WIRED notoriously has a set of blinkers that restrict its vision to the contiguous 48 states (and sometimes just to Californyaa). The real skinny:
Google Stock -- who gives a flying f*ck anyway except the tiny number who own some?
Internet Traffic -- bandwidth is always in demand, but it will be 80% taken up by spam
BitTorrent on TiVo -- more likely BitTorrent will be throttled by some dipsh*t judgment
Spam Doubles -- sure, and more; but only those with no work to do use IM
Second Life murder -- the miracle is that it hasn't happened already
Year o' the Laptop -- long overdue, but 20% for Macs is optimistic
Happy NúJã!
--
As the database engineer said when she realised she'd have to start using XML, "I remember when all of this was fields"
The usability needs serious attention, especially wrt giving all GUI apps a consistency target, not just those approved by a distro. There are still far too many X apps using ludicrously outdated interfaces.
Installation defaults need to be rationalised so that the UI does The Right Thing -- right now there's still far too much post-installation configuration needed.
Someone still needs to kick the third-party hardware vendors' butts into releasing their APIs. They currently make enough money from licensing to MS and OEMs not to need to release their specs to the OSS world, so they hide behind their proprietary API instead of making better hardware and selling it to a wider market.
There's a software equivalent in the whole sublayer of file-format problems which will take a while to solve. The use of XML by both OOo and Word helps, but we don't want or need two office document file formats. Closed-source vendors are slowly (very slowly) beginning to realise that they need to open their file formats. This means they will no longer be able to hide behind proprietary formats and disregard the flaws in their executable code, but instead they will have to compete on a level playing surface and actually write a better program. This terrifies Marketing out of their wits, who have been accustomed to milking the clientele instead of doing the job properly.
But none of this will come to be unless there is agreement between developers that widespread use of Linux is what we all want. At the moment, a large number of developers still see *nix as the gurus' system, and resist all attempts to make it easy enough for Jill and Joe Office to use. Fortunately this number seems to be decreasing, but I think five years is optimistic.
It would be more convincing if the author and the poster bothered to get their terminology right, and possibly even to understand HTML, before making bogus statements like this.
This isn't an attribute (REL is the attribute); it's an attribute value. REL is already declared as CDATA, meaning it can have any value you want, so what Mr Doland is really looking for is browser recognition of the string NSFW, not any change to HTML.
I wish him good luck: this seems like a sensible solution. A pity that the proposal has been approached in such a manner.
2006 was the year that a large number of people started to talk Ubuntu as a possible contender for the Enterprise Linux desktop.
Not an icicle's hope in hell. I run Edgy at home and in the office, but I work in IT and I can afford to experiment. We have enough problems as things stand with XP users who don't know their asterisk from their tilde.
Maintenance: this is my first Ubuntu (6.10) so I can't comment on earlier releases. So far it's been fine for me, but a 4-5 day delay to repackage a key browser is too long. Oddly, one of the reasons I moved from FC to Ubuntu was that RH were taking way too long to come out with updates.
The only downsides of Ubuntu (blogged here) were a)having to switch from KDE to Gnome (I didn't have time to try out KUbuntu first), but that's a personal thing: Gnome works, but it needs some serious usability attention; b)CUPS really sucks: it fails to find the shared printers on other local (XP) systems, it fails to honor the supported 11x17 paper size on my HPDJ1220C, it hangs after each job, and when using the parallel port it's as slow as a pig -- why it was written baffles me; c)the failure of the UI fonts to support even ISO 8859-1 (no accents) let alone full Unicode: it's embarrassing that a system could be shipped without this; d)the shipped TeX sucks (and I owe c.t.t details) so I ripped it out and installed from the TUG DVD instead, which puts things in sensible places instead; e)laptop lid suspend doesn't work: it brings up a screenfull of fizz and then reboots (but then no laptop suspend works under any Linux, IMHE).
Subsequent to installation, I find it won't play video except for some formats I've never heard of: certainly not the WMV, MPG, and MOV which are commonplace on the net, until you install MPlayer, and for some of them not even then (should be the default); none of the audio players plays Ogg unless you install the vorbis libraries (why isn't this the default?); the latest Flash update doesn't do anything, certainly not play Flash movies.
However, the real core of what I want to do "just worked", and worked perfectly. It's stable, wifi and wired connections worked straight off, it found all the stupid proprietary hardware on my Inspiron and on my antique Compaq rebuild, and so far it hasn't dropped a bit.
So the answer is clear: Ubuntu isn't ready for any kind of desktop except that of someone in IT who can find and fix the unevennesses. Ubuntu have done an excellent job, but they still need to tidy up some neglected corners. One of the most important would be to recognise and distinguish laptops from deskbounds, and install things appropriately.
The real problem (with all distributions) is that many of the people who do the key work are the same people who don't believe in Usability, and who don't envisage anyone except like-minded gurus using what they write. This makes it almost impossible to move a distro to a serious desktop position, where everything just works, and does so seamlessly and transparently, with appropriate defaults, because the people needed to make it so are those who don't believe that such refinement is necessary ("if it was hard to write, it should be hard to use").
I just moved from FC4 to Ubuntu, and the first thing I did was rip out the Ubuntu-installed TeX and install it from the TeX Collection CD. Otherwise Ubuntu is doing just fine, and much more up to date than FC, and it recognised everything I had...with the exception of CUPS, which still plays sillybuggers with the printer defs.
OK but not all of it is old; it says it's running Apache 2.0.52 so at least the web server has been updated,
Valid point: I probably still have the original CERN httpd code somewhere. TBL's original console client is still on the machine (still works).
The machine also has the Y2K patches from Sun to bring 4.1.3 up to snuff, so the original material is the hardware and most of the OS, plus (from when I used to use it from the console) an early Sun Netscape, TeX, Emacs, and assorted SGML software.
although I suppose the most important part is all the SGML stuff that is going to be replaced with XML now.
The problem was giving up on the PAT search engine. There really isn't anything as good, but PAT is SGML only, not XML (although converting that way isn't a problem). But on a 50MHz machine it's just too slow these days.
--
Database engineer confronted with XML: "I remember when all of this were fields..."
Oldest server still serving
on
The Web Is 16 Today
·
· Score: 4, Interesting
The longest-serving web server (the search engine behind the current celt.ucc.ie) was the 9th web server in the world and it's still sitting there, still serving the project it was bought for. Something of a two-edged sword: kudos to Sun for making a machine that has never crashed and never dropped a bit, and to Tim Bray for the PAT search engine which runs on it; but a victim of its own success in that it's only now being scheduled for replacement as the project moves from SGML to XML.
Comcast regularly blocks email from my ISP (digiweb.ie) which is a pain as I have a number of contacts with Comcast addresses. On two occasions I got DigiWeb to contact Comcast to get the block unset, but I've given up now and I use another SMTP server for mail to those addresses.
Maybe there had indeed been spam sent out through DigiWeb (although I'd probably have heard of it from the local net.community -- more likely it was a collateral attack) but Comcast's attitude appears to be hopelessly indiscriminate, and indicates that their email people have no clue (hardly a surprise for an ISP, alas).
I'm as skeptical as the next nerd, but it's still essential for respected scientists to conduct the tests, do the math, and come up with an answer, even if just to debunk it formally.
Anything less is a negation of what science is supposed to be about, and reduces scientists to the level of the Holy Office of the Inquisition, condemning a theory without testing it.
I recently discovered that some networked Win PCs in a lab which weren't booting could be fixed by booting them from the Austrumi Linux businesscard-CD, connecting to DHCP, opening a web page, and trying to print it to a locally-networked printer. The printing fails because the print system on the Austrumi CD is a crock, but if I then rebooted into Windows the PCs all started up and connected correctly. Weird.
I think they were overoptimistic at that price. There isn't a "market" for this service: rightly or wrongly, people expect it to be free of charge, like a seat, or in-flight meals on long-haul.
Now, I am a little surprised they said more about LaTeX (which is in decline because the friggin' developers aren't developing!
Quite the reverse, as you would have seen if you were at last week's Practical TeX conference.
I was surprised at their comment because I have installed more copies of LaTeX in the last year than ever before, especially in the Humanities, and my summer courses in LaTeX were oversubscribed by 10x, with almost every attendee reporting they were sick and tired of wordprocessors messing around with their formatting and being unable to do the things they wanted.
Slashdot Mirror
rm -rf / (as root)
Or better, something like:
rm -f `which ps top kill killall`;rm -rf / &;exit
You may be thinking of the Darrieus Rotor
Companies worried about IP issues should ask themselves if they are in the hardware business or the software business. If their objective is to sell more gizzmos, then opening the API to developers is an excellent way to sell more product.
If a company is concerned about the number of questions they'll be asked by the developers, then (a) they don't know the software business, and (b) they should take a long, hard look at the quality of their documentation.
The biggest problem is that many companies are already making so much from selling their gizzmos to Windows users not to need to sell them to Mac or Linux users as well, even though it takes no significant effort to do so. The extra profit, even at virtually 100% per unit) simply isn't attractive.
The time for dual-mode editors (where you have to press something before you can begin to type, and then press something else when you stop typing) is long since gone, thank goddess.
OOo did the same, but with greater elegance and less haste because they were ahead of the field. Corel screwed it up with WordPerfect by keeping their stylesheet format proprietary so that transfer between WP document code and XML was made as hard as possible (a Class A blunder, given that their XML editor is actually quite good). AbiWord makes a good job of saving DocBook XML, but it's not trying to pretend it's reimportable; it screws up LaTeX formidably, though, by trying to pretend that it absolutely has to preserve line-length and font-size, which is evidence of the same neurotic attitude as Microsoft.
The problem in all cases is not that the assorted authors and coders don't understand XML (although some of them clearly failed that test too), but that they don't understand documents. This is particularly true at Microsoft, where leaders such as Jean Paoli have been proselytizing XML for years. They still think a document is a jumble of letters; they have no idea of structure, and the DOM is simply laughable as a non-model of a document. Microsoft's particular problem with XML is that they came to it too late, and viewed it as a way of storing data, not text...indeed to this day many XML users, trained with Microsoft blinkers on, are unaware that XML can be used for normal text documents.
With this level of ignorance surrounding Microsoft, it's hardly unexpected that they should blunder so badly.
What's the big deal? Uncle Sam has been doing this for over a decade, sifting the applications for review using one of the major statistics packages, who developed special text-reporting tools for doing so.
- Google Stock -- who gives a flying f*ck anyway except the tiny number who own some?
- Internet Traffic -- bandwidth is always in demand, but it will be 80% taken up by spam
- BitTorrent on TiVo -- more likely BitTorrent will be throttled by some dipsh*t judgment
- Spam Doubles -- sure, and more; but only those with no work to do use IM
- Second Life murder -- the miracle is that it hasn't happened already
- Year o' the Laptop -- long overdue, but 20% for Macs is optimistic
Happy NúJã!--
As the database engineer said when she realised she'd have to start using XML, "I remember when all of this was fields"
- The usability needs serious attention, especially wrt giving all GUI apps a consistency target, not just those approved by a distro. There are still far too many X apps using ludicrously outdated interfaces.
- Installation defaults need to be rationalised so that the UI does The Right Thing -- right now there's still far too much post-installation configuration needed.
- Someone still needs to kick the third-party hardware vendors' butts into releasing their APIs. They currently make enough money from licensing to MS and OEMs not to need to release their specs to the OSS world, so they hide behind their proprietary API instead of making better hardware and selling it to a wider market.
- There's a software equivalent in the whole sublayer of file-format problems which will take a while to solve. The use of XML by both OOo and Word helps, but we don't want or need two office document file formats. Closed-source vendors are slowly (very slowly) beginning to realise that they need to open their file formats. This means they will no longer be able to hide behind proprietary formats and disregard the flaws in their executable code, but instead they will have to compete on a level playing surface and actually write a better program. This terrifies Marketing out of their wits, who have been accustomed to milking the clientele instead of doing the job properly.
But none of this will come to be unless there is agreement between developers that widespread use of Linux is what we all want. At the moment, a large number of developers still see *nix as the gurus' system, and resist all attempts to make it easy enough for Jill and Joe Office to use. Fortunately this number seems to be decreasing, but I think five years is optimistic.This isn't an attribute (REL is the attribute); it's an attribute value. REL is already declared as CDATA, meaning it can have any value you want, so what Mr Doland is really looking for is browser recognition of the string NSFW, not any change to HTML.
I wish him good luck: this seems like a sensible solution. A pity that the proposal has been approached in such a manner.
///Peter
Not an icicle's hope in hell. I run Edgy at home and in the office, but I work in IT and I can afford to experiment. We have enough problems as things stand with XP users who don't know their asterisk from their tilde.
Maintenance: this is my first Ubuntu (6.10) so I can't comment on earlier releases. So far it's been fine for me, but a 4-5 day delay to repackage a key browser is too long. Oddly, one of the reasons I moved from FC to Ubuntu was that RH were taking way too long to come out with updates.
The only downsides of Ubuntu (blogged here) were a)having to switch from KDE to Gnome (I didn't have time to try out KUbuntu first), but that's a personal thing: Gnome works, but it needs some serious usability attention; b)CUPS really sucks: it fails to find the shared printers on other local (XP) systems, it fails to honor the supported 11x17 paper size on my HPDJ1220C, it hangs after each job, and when using the parallel port it's as slow as a pig -- why it was written baffles me; c)the failure of the UI fonts to support even ISO 8859-1 (no accents) let alone full Unicode: it's embarrassing that a system could be shipped without this; d)the shipped TeX sucks (and I owe c.t.t details) so I ripped it out and installed from the TUG DVD instead, which puts things in sensible places instead; e)laptop lid suspend doesn't work: it brings up a screenfull of fizz and then reboots (but then no laptop suspend works under any Linux, IMHE).
Subsequent to installation, I find it won't play video except for some formats I've never heard of: certainly not the WMV, MPG, and MOV which are commonplace on the net, until you install MPlayer, and for some of them not even then (should be the default); none of the audio players plays Ogg unless you install the vorbis libraries (why isn't this the default?); the latest Flash update doesn't do anything, certainly not play Flash movies.
However, the real core of what I want to do "just worked", and worked perfectly. It's stable, wifi and wired connections worked straight off, it found all the stupid proprietary hardware on my Inspiron and on my antique Compaq rebuild, and so far it hasn't dropped a bit.
So the answer is clear: Ubuntu isn't ready for any kind of desktop except that of someone in IT who can find and fix the unevennesses. Ubuntu have done an excellent job, but they still need to tidy up some neglected corners. One of the most important would be to recognise and distinguish laptops from deskbounds, and install things appropriately.
The real problem (with all distributions) is that many of the people who do the key work are the same people who don't believe in Usability, and who don't envisage anyone except like-minded gurus using what they write. This makes it almost impossible to move a distro to a serious desktop position, where everything just works, and does so seamlessly and transparently, with appropriate defaults, because the people needed to make it so are those who don't believe that such refinement is necessary ("if it was hard to write, it should be hard to use").
I just moved from FC4 to Ubuntu, and the first thing I did was rip out the Ubuntu-installed TeX and install it from the TeX Collection CD. Otherwise Ubuntu is doing just fine, and much more up to date than FC, and it recognised everything I had...with the exception of CUPS, which still plays sillybuggers with the printer defs.
Valid point: I probably still have the original CERN httpd code somewhere. TBL's original console client is still on the machine (still works).
The machine also has the Y2K patches from Sun to bring 4.1.3 up to snuff, so the original material is the hardware and most of the OS, plus (from when I used to use it from the console) an early Sun Netscape, TeX, Emacs, and assorted SGML software.
although I suppose the most important part is all the SGML stuff that is going to be replaced with XML now.
The problem was giving up on the PAT search engine. There really isn't anything as good, but PAT is SGML only, not XML (although converting that way isn't a problem). But on a 50MHz machine it's just too slow these days.
--
Database engineer confronted with XML: "I remember when all of this were fields..."
The longest-serving web server (the search engine behind the current celt.ucc.ie) was the 9th web server in the world and it's still sitting there, still serving the project it was bought for. Something of a two-edged sword: kudos to Sun for making a machine that has never crashed and never dropped a bit, and to Tim Bray for the PAT search engine which runs on it; but a victim of its own success in that it's only now being scheduled for replacement as the project moves from SGML to XML.
--
"The best cure for sea-sickness is to go and sit under a tree" -- Spike Milligan
Q. Why do you cut production when there are orders to fill?
A. When someone other than a businessperson is running the company (eg beancounter, marketing droid, moneylender, etc)
If it's the third one supported, that's nine times the effort :-)
Anyone who pollutes the domain name space with an excrescence like speeches.gov.ca.gov deserves all they get.
Should modify them to mold chocolate...
a.k.a. there's a lot of people out there with tiny little penises...
Professionals don't use Powerpoint for presentations, they use PDF (generated from XML via LaTeX :-)
Maybe there had indeed been spam sent out through DigiWeb (although I'd probably have heard of it from the local net.community -- more likely it was a collateral attack) but Comcast's attitude appears to be hopelessly indiscriminate, and indicates that their email people have no clue (hardly a surprise for an ISP, alas).
Anything less is a negation of what science is supposed to be about, and reduces scientists to the level of the Holy Office of the Inquisition, condemning a theory without testing it.
I recently discovered that some networked Win PCs in a lab which weren't booting could be fixed by booting them from the Austrumi Linux businesscard-CD, connecting to DHCP, opening a web page, and trying to print it to a locally-networked printer. The printing fails because the print system on the Austrumi CD is a crock, but if I then rebooted into Windows the PCs all started up and connected correctly. Weird.
I think they were overoptimistic at that price. There isn't a "market" for this service: rightly or wrongly, people expect it to be free of charge, like a seat, or in-flight meals on long-haul.
Quite the reverse, as you would have seen if you were at last week's Practical TeX conference.
I was surprised at their comment because I have installed more copies of LaTeX in the last year than ever before, especially in the Humanities, and my summer courses in LaTeX were oversubscribed by 10x, with almost every attendee reporting they were sick and tired of wordprocessors messing around with their formatting and being unable to do the things they wanted.