You are half-right. A standard widget, and component library are needed, but I disagree with the need to integrate it with X.
X is a network protocol. If you look at network protocol stacks you see layered design patterns everywhere. That's the beauty of X. It is confined to a layer, and performs that layer service extremely well. I know it has drawbacks and inefficiencies, but it is the best protocol so far.
Including widgets and component architectures is stepping on the upper layer. It violates layer independency, and introduces unneeded complexity. X is a large enough behemoth as it is.
Leave widgets separate, as they are now. It works, and it is elegant.
BTW, this is also why I disagree with the various alternatives to X that discard the network protocol, and go for direct hardware communication. It is a decision that mixes the graphical communication layer with the layer beneath it -- you gain some speed and loose lots of flexibility.
The two main desktop environments for Linux (KDE and Gnome) do not offer usability improvements over mainstream desktop Operating Systems -- namely MS Windows and MacOS.
Gnome does have a usability project. What is your opinion on its actual impact on Gnome? Do you feel the open-source movement can attract non-programmers -- like usability experts -- with the same intensity it attracts programmers?
... learn this basic rule-of-thumb: "Never, ever, act against you customers."
Amazon provides an excellent service for the reading community, and therefore to publishers and authors. No, it doesn't lick publishers' boots like authors are forced to. And yes, it is big enough that it poses a menace to publishers -- it is big enough to exert a visible pressure on the market.
Acting against Amazon is stupid. It is futile -- the market chooses the best service, and doesn't give a shit about the publishers. It is also stupid, from a publishers' view. Destroying Amazon would decrease book sales overall. Isn't this obvious?!
This goes in line with movie producers wanting me to use a XPTO-approved device for viewing movies. Expected result -> I won't buy movies -> less business overall.
I really really hate AES -- Analogy Exageration Syndrome. Whenever you make an analogy, take the time to see if you are introducing noise in your line of reasoning.
Case in question:
Blind people are not disabled by option. Except for Bob the dinossaur, people don't go around sticking pencils in their eyes. It is obviously impossible to make them responsible for something they do not control (whether they can see the ads or not).
A 73Gb SCSI disk from IBM goes for a thousand dollars. That gives 13 USD per gigabyte, and a measly 13 cents for your whole 10 megs of crap.
I'd say they won't give a damn if you keep your data there or not.
The real issue here is evaluating if their service is good enough to be paid for. Ads don't cut it, never did, and don't promise to provide enough revenue in the future - like they did last year. The only other source of income are customers. If the service is good enough, pay for it. If it isn't buy elsewhere.
IBM has just surpassed Sun in server sales, and sells pretty good Linux boxes. I have a datacenter of IBM xSeries servers, and apart from a severe disk problem, now gone, I find IBM servers fantastic.
The top two players in the server market sell Unix based solutions. Dell is playing M$'s game, hoping to ride the Microsoft rollercoaster to the top. It *is* a smart move, since the alternative involves doing some real effort to provide some real service.
I don't know how is Dell in the US. Here (Portugal), I asked for a quote on their site (when purchasing the datacenter hardware). They took the better part of two weeks to answer, sending me a proposal in MS Word format and written in Spanish. No excuse for the delay, and no excuse for not using my native language or English. By that time, I had narrowed negotiations to IBM and Sun, and was closing contract with IBM. I dropped Dell, didn't answer them, and overall came out with a very bad impression of their service.
... that I'll keep sticking to my ol'faithfull windowmaker.
Seriously, though, with De Icaza chasing MS tail lights on one side, and KDE creating monolitic frameworks -- kcalc has the biggest footprint I've ever seen for a calculator -- on the other, I fail to see anything appealing with the current linux desktop solutions.
Re:Those opening paragraphs...
on
.NETly News
·
· Score: 1
Previous Post
Does anyone truly believe that Gates has made a positive contribution to "this earth", other than his (admittedly laudable) charitable works?
Your post
I don't consider 20+ billion dollars in one shot 'admittedly laughable'.
From everything2.com:
Laudable: Praiseworthy, or commendable, such as laudable motives and actions. Laughable: Fitted to excite laughter; as, a laughable story; a laughable scene.
A big difference, isn't it?
Anyhow, without devaluing Gates' charity ideals, money is largely evaluated on the basis of how much time one takes to gather a given amount. I bet Gates makes 20 million in about the same time I need to write this post... 'nough said!
The point of it is that even if (well, rather when, in my opinion) Microsoft starts corrupting the standard it doesnt really matter because the point isnt really interoperability with MS, it is getting another useful development platform that isnt as tied to languages as the ones we have now.
I don't get it. If the point is not interoperability, and is just developing a cool-platform-of-the-day, why not design it from scratch?. Lots of stuff could be taken from other virtual-machine based projects, like blackdown or python.
This just ties MS with open-source good press -- which is annoying -- and ties open-source with MS's bad press -- which is unacceptable.
Gnome is through for me. Heck, I came back to good ol' windowmaker six months ago. It's a shame to see a good open-source project go down the drain, because of a narrow minded leader.
Microsoft may reliquish desktop OS monopoly. It will not give up internet servers/protocols easily. Desktop OSs will probably become irrelevant in the future. Network communications are the next money-making gateway.
That could be true, were not Peter Jackson one of my favourite directors of all time, and were not all of his movies I've seen B-grade movies. Budgets so low that in "Braindead" they used pork's blood from a meat shop instead of artificial blood, in the lawn mowner scene.
Please, try to see "Meet the Feebles" and "Braindead" to get a good grip on Jackson's style. "Bad Taste" is also a good one, following Braindead's style. "Meet the Feebles" got the theater I was in laughing for one hour straight. Absolutely amazing.
Re:One simple reason why it won't work:
on
The Euro
·
· Score: 2, Insightful
I can't confirm your de facto barriers for mobility. Language is certainly no problem, and cultural difference is small enough to be an incentive rather than a deterrent.
I'm Portuguese, and I can say that Portuguese usually speak two foreign languages -- English and French. Couple that with the language proximity to Italian and Spanish (enough to get you started, if you travel there), and you've opened up a great part of Europe.
I'd say that Europeans tend to settle when they get older (mid-thirties). But that's natural. It's impossible for 100% of the population to be mobile. Younger people tend to grab the oportunity to travel. I don't know the US, but I'd guess the average red-neck Texan won't move to Sillicon Valley too quickly either,
You obviously don't know WM. Windowmaker's version numbers grow veeeeeery slowly. True open-source fashion. Good stuff takes time to get right. Just like good wine.:-)
And for all of you out there, who have also never tried windowmaker. Go, and take it out for a spin. Use it on your productivity desktop for a week. It's diferent, it takes time to get used to, but it pays. It never, ever stands in your way. You don't realise it's there, until you start something else by mistake and you miss yer'ol wm. That's the signature of good things!
What I just can't stand in Opera is the damned MDI window management. That has been abandoned for years, and for good reason.
People invest lots of hours creating a good window manager, just for some stupid app to go and declare it wants to manage its own windows. Simply stupid.
I do agree that it is fast and pretty standards compatible. I stick with the lizzard, though. As for other people, I don't care.
What I really would like to see is the ability to have Relational Databases, with hierarchical types for fields. I would be able to query these fields much like I query/transform an XML document, possibly using the XML technologies (XPath, XSL,...)
The relational model is very good for most situations, and has been very studied and optimized. Noone would transition back to pure hierarchical DBs.
You could say that. But I must point out that 80-85 is my limit. It usually runs below that point, not usually above as the quote mentioned.
My personal experience with ReiserFS is excellent, including running a 12 Million pageviews a month site, with some havy database serving.
I just don't understand all the fuss around ext3. It's just-another-journaling-filesystem. We've had those for some months now... And I certainly don't tolerate FUD applied to kernel politics.
Just treat ext3 just like one more journaling filesystem. This one has easy conversion, and that's just about it. Don't bash on ReiserFS stability, because today, it's a non-issue. It's codebase is more tested than ext3's.
You could also use GRUB to boot, which reads ReiserFS and probably ext3 (at least as an ext2). Also, if you mount it read-only (sufficient for everyday use), there are no fsck involved, even for ext2.
...but slows down as the filesystem approaches 90% usage (which is pretty typical for a production box)
If he said 90% disk usage is typical for a production box, he was wrong. I never leave a server of mine above 80%-85% disk usage. Not unless I'd love 3am calls to clean up disk space for some fucking temp file. And not at today's hd prices.
When you first start a fresh install of Window Maker, you get a nearly-blank screen with a dock in the corner.
I've not installed windowmaker from sources in a long time, so I don't know how does it come out of a source install. I can tell you that with SuSE, it comes preinstalled with a good menu you probably don't need to edit, it comes with a few dockapps preconfigured, and with one windows-user-friendly dockapp which opens the menu, and placed at the bottom-left.
Usability has two definitions:
a. Is easy to use, even for unexperienced people.
b. Maximizes user potential, by allowing users to do what they want to do, as fast as possible.
Windowmaker fits b. for me, and for lots of people. No UI besides windows will fit a. for a windows user, because computer-human interaction always requires learning.
wow! Talk about being vague. Some counterexamples:
Usability: Windowmaker, taking the best from the labs at now defunct NeXT. Very very good.
Aesthetics: huh? This is highly personal, but some enlightenment themes are breathtaking.
Integration: StarOffice. Open, documented, plain-text document formats, unlike M$ Word.doc which must be reverse-engineered for every release.
Feature Completeness: You must look at stable projects. LaTeX, Emacs or Windowmaker are good examples. Unlike commercial bloatware, OSS does not need to continuosly evolve, and tends to stabilize when needs are fullfilled,
Support: I grant you this one, *if* your notion of support is having someone to call 24/7 and to blame if s/w crashes.
Documentation: Take a look at GTK or Qt documentation. Clear, and thorough.
Stability: Get a stable debian, don't use bleeding edge, pre-1.0 alfa versions.
Ease of instalation: SuSE 7.2. Damn fast install, all questions asked at the very beggining and so easy that my mom could install it.
Hardware Support: My TV capture card (Bt-based). Works MUCH better with Linux drivers than buggy windows ones. Lack of h/w support was a problem two years ago. Now, with most h/w companies embracing OSS, it's better than in Windows.
Availability of s/w: Depends on what you mean. You won't find any good desktop publishing app, for instance (only on MacOS), but you'll find dozens of web servers.
TCO: Big industry-pushed lie. Ex: Patch a campus-wide network of NTs with the latest service patch: it takes days. Now, do an automatic Yast online update on the same network: about 20min.
As for the apps, its a never ending discussion. Gcc is much better than VC++, KDE is more than ready for prime-time, I only use GIMP for my web image editing needs, for instance.
If he's not able to make himself socially presentable for an interview, don't hire him.
Why should I hire a candidate, who doesn't grasp the simple rules by which the hiring-game is played?
Chicken and egg problem... Why should the candidate let himself be hired by someone who spends precious interview time with "hiring games" of little relevance to the everyday task?
I don't wear suit and tie for interviews, as much as I don't wear suit and tie in my everyday job. My presentation in the interview is as close as possible to what my employer will have to live with. Never failed a job...
The Windows world is not the *nix world. People don't wait for the.1 or.2 release, they expect the.0 releases to work as they should.
Ha! Yeah, how many people do you know who have really used Windows 1.0, Word 1.0, Excel 1.0 or Windows CE 1.0? If there's a company which never gets it right the first time, it's Microsoft. I'm willing to bet that the Xbox will crash often (console crashes are very rare, you know).
Slashdot Mirror
SuSE evaluation CDs. They run off the CD, sport a KDE desktop, and are great for showing off Linux.
X is a network protocol. If you look at network protocol stacks you see layered design patterns everywhere. That's the beauty of X. It is confined to a layer, and performs that layer service extremely well. I know it has drawbacks and inefficiencies, but it is the best protocol so far.
Including widgets and component architectures is stepping on the upper layer. It violates layer independency, and introduces unneeded complexity. X is a large enough behemoth as it is.
Leave widgets separate, as they are now. It works, and it is elegant.
BTW, this is also why I disagree with the various alternatives to X that discard the network protocol, and go for direct hardware communication. It is a decision that mixes the graphical communication layer with the layer beneath it -- you gain some speed and loose lots of flexibility.
Gnome does have a usability project. What is your opinion on its actual impact on Gnome? Do you feel the open-source movement can attract non-programmers -- like usability experts -- with the same intensity it attracts programmers?
The problem with hardware is that it actually costs a lot of money to keep the development versions in the market. Let's wait and see.
Amazon provides an excellent service for the reading community, and therefore to publishers and authors. No, it doesn't lick publishers' boots like authors are forced to. And yes, it is big enough that it poses a menace to publishers -- it is big enough to exert a visible pressure on the market.
Acting against Amazon is stupid. It is futile -- the market chooses the best service, and doesn't give a shit about the publishers. It is also stupid, from a publishers' view. Destroying Amazon would decrease book sales overall. Isn't this obvious?!
This goes in line with movie producers wanting me to use a XPTO-approved device for viewing movies. Expected result -> I won't buy movies -> less business overall.
Case in question:
Blind people are not disabled by option. Except for Bob the dinossaur, people don't go around sticking pencils in their eyes. It is obviously impossible to make them responsible for something they do not control (whether they can see the ads or not).
When the density of "thing" thingies on any on of my phrase gets above a certain limit, it's a sign that it is a good idea to thing^Hk before posting.
I'd say they won't give a damn if you keep your data there or not.
The real issue here is evaluating if their service is good enough to be paid for. Ads don't cut it, never did, and don't promise to provide enough revenue in the future - like they did last year. The only other source of income are customers. If the service is good enough, pay for it. If it isn't buy elsewhere.
The top two players in the server market sell Unix based solutions. Dell is playing M$'s game, hoping to ride the Microsoft rollercoaster to the top. It *is* a smart move, since the alternative involves doing some real effort to provide some real service.
I don't know how is Dell in the US. Here (Portugal), I asked for a quote on their site (when purchasing the datacenter hardware). They took the better part of two weeks to answer, sending me a proposal in MS Word format and written in Spanish. No excuse for the delay, and no excuse for not using my native language or English. By that time, I had narrowed negotiations to IBM and Sun, and was closing contract with IBM. I dropped Dell, didn't answer them, and overall came out with a very bad impression of their service.
Seriously, though, with De Icaza chasing MS tail lights on one side, and KDE creating monolitic frameworks -- kcalc has the biggest footprint I've ever seen for a calculator -- on the other, I fail to see anything appealing with the current linux desktop solutions.
Laudable: Praiseworthy, or commendable, such as laudable motives and actions.
Laughable: Fitted to excite laughter; as, a laughable story; a laughable scene.
A big difference, isn't it?
Anyhow, without devaluing Gates' charity ideals, money is largely evaluated on the basis of how much time one takes to gather a given amount. I bet Gates makes 20 million in about the same time I need to write this post... 'nough said!
This just ties MS with open-source good press -- which is annoying -- and ties open-source with MS's bad press -- which is unacceptable.
Gnome is through for me. Heck, I came back to good ol' windowmaker six months ago. It's a shame to see a good open-source project go down the drain, because of a narrow minded leader.
Microsoft may reliquish desktop OS monopoly. It will not give up internet servers/protocols easily. Desktop OSs will probably become irrelevant in the future. Network communications are the next money-making gateway.
-
Compaq Proliant DL590
-
Linux MM homepage Look at the 64Gb entry from 1999
I'm in no way defending either side, but we must be correct about facts.I do use Linux/Intel in a budget conscious organization, and so far the xSeries servers from IBM have behaved like a charm. I've never deployed SUNs.
Please, try to see "Meet the Feebles" and "Braindead" to get a good grip on Jackson's style. "Bad Taste" is also a good one, following Braindead's style. "Meet the Feebles" got the theater I was in laughing for one hour straight. Absolutely amazing.
I'm Portuguese, and I can say that Portuguese usually speak two foreign languages -- English and French. Couple that with the language proximity to Italian and Spanish (enough to get you started, if you travel there), and you've opened up a great part of Europe.
I'd say that Europeans tend to settle when they get older (mid-thirties). But that's natural. It's impossible for 100% of the population to be mobile. Younger people tend to grab the oportunity to travel. I don't know the US, but I'd guess the average red-neck Texan won't move to Sillicon Valley too quickly either,
And for all of you out there, who have also never tried windowmaker. Go, and take it out for a spin. Use it on your productivity desktop for a week. It's diferent, it takes time to get used to, but it pays. It never, ever stands in your way. You don't realise it's there, until you start something else by mistake and you miss yer'ol wm. That's the signature of good things!
People invest lots of hours creating a good window manager, just for some stupid app to go and declare it wants to manage its own windows. Simply stupid.
I do agree that it is fast and pretty standards compatible. I stick with the lizzard, though. As for other people, I don't care.
The relational model is very good for most situations, and has been very studied and optimized. Noone would transition back to pure hierarchical DBs.
My personal experience with ReiserFS is excellent, including running a 12 Million pageviews a month site, with some havy database serving.
I just don't understand all the fuss around ext3. It's just-another-journaling-filesystem. We've had those for some months now... And I certainly don't tolerate FUD applied to kernel politics.
Just treat ext3 just like one more journaling filesystem. This one has easy conversion, and that's just about it. Don't bash on ReiserFS stability, because today, it's a non-issue. It's codebase is more tested than ext3's.
You could also use GRUB to boot, which reads ReiserFS and probably ext3 (at least as an ext2). Also, if you mount it read-only (sufficient for everyday use), there are no fsck involved, even for ext2.
If he said 90% disk usage is typical for a production box, he was wrong. I never leave a server of mine above 80%-85% disk usage. Not unless I'd love 3am calls to clean up disk space for some fucking temp file. And not at today's hd prices.
Usability has two definitions:
a. Is easy to use, even for unexperienced people.
b. Maximizes user potential, by allowing users to do what they want to do, as fast as possible.
Windowmaker fits b. for me, and for lots of people. No UI besides windows will fit a. for a windows user, because computer-human interaction always requires learning.
- Usability: Windowmaker, taking the best from the labs at now defunct NeXT. Very very good.
-
Aesthetics: huh? This is highly personal, but some enlightenment themes are breathtaking.
-
Integration: StarOffice. Open, documented, plain-text document formats, unlike M$ Word
.doc which must be reverse-engineered for every release.
-
Feature Completeness: You must look at stable projects. LaTeX, Emacs or Windowmaker are good examples. Unlike commercial bloatware, OSS does not need to continuosly evolve, and tends to stabilize when needs are fullfilled,
-
Support: I grant you this one, *if* your notion of support is having someone to call 24/7 and to blame if s/w crashes.
-
Documentation: Take a look at GTK or Qt documentation. Clear, and thorough.
-
Stability: Get a stable debian, don't use bleeding edge, pre-1.0 alfa versions.
-
Ease of instalation: SuSE 7.2. Damn fast install, all questions asked at the very beggining and so easy that my mom could install it.
- Hardware Support: My TV capture card (Bt-based). Works MUCH better with Linux drivers than buggy windows ones. Lack of h/w support was a problem two years ago. Now, with most h/w companies embracing OSS, it's better than in Windows.
-
Availability of s/w: Depends on what you mean. You won't find any good desktop publishing app, for instance (only on MacOS), but you'll find dozens of web servers.
-
TCO: Big industry-pushed lie. Ex: Patch a campus-wide network of NTs with the latest service patch: it takes days. Now, do an automatic Yast online update on the same network: about 20min.
As for the apps, its a never ending discussion. Gcc is much better than VC++, KDE is more than ready for prime-time, I only use GIMP for my web image editing needs, for instance.I don't wear suit and tie for interviews, as much as I don't wear suit and tie in my everyday job. My presentation in the interview is as close as possible to what my employer will have to live with. Never failed a job...