I can't see how did I blame Microsoft monopoly. I did mean that, Palm keeping its current strategy of selling Dragonball-based Palms, then iPaq and family will win the market.
I am a Palm III and Palm Vx owner. I carried a Palm for the last ~4 years. Its a good product, but one must realize that this is the computer industry. One doesn't sit tight on a good product for five years.
PPCs on the other hand, are horrendous devices. MS didn't resist on slapping the desktop GUI on those things. How dumber can you go over adding a taskbar on such a small screen?!
I believe PPCs are inferiour products. For equal evolution, Palm would give a much better platform (yes, I've programmed both PalmOS and PPCs). Monopoly has nothing to do with it.
They had one hell of a product, back in 1997. Yes, that's five years ago!!!
Product development since then? Zero, zip, niente, nada de nada. They let go all of their competent techies, and are now a mass of marketeers without guindance, slowly sinking to the sound of Titanic's band.
It's really sad that these guys took Psion's market, and then managed to give it away to M$.
If you really want to unify the various pipes into one big fat pipe, you'll need software on both ends. The software can act on any layer of the network stack, and unify the bandwidth for the layers above.
This would tipically be done on the data-link layer, but can also be done at the IP level.
If you can't change the software on the other end, you'd have to do the unification at the application layer. It depends on the application. Ex: for HTTP, you can balance the requests over the lines. When the size is known, you can do partial requests and balance these along the lines; for FTP, do PART requests for the pieces of the file.
Expose a proxy for each application protocol. Not all applications can be adapted to take advantage of the multiplexing, however.
... of a national ID number. In Europe its quite common -- except for the brits. The real question is whether companies or the state can do joins on the different databases.
Portuguese laws forbid different entities from cross-referencing their databases, without explicit approval of the citizens. The way it is written, it even affects different departments of the state -- leading to a social security number, a tax ID number, etc.
I think it is a lot worse the way its done in the US, where everyone and their dog knows your SS#. It is very easy to cross-reference the DBs. At least here, they'd need to do some data mining...
It just depends on *how* insecure it really was. If it was really bad, driving around with a wireless-enabled laptop running XP could result in a five-year jail sentence. With XP's automatic wireless lan setup and all.
His biggest error probably was talking about it. He should have sold the info to some mobster gang. They'd probably be much more gratefull.
Your ISP didn't necessarily sell his user list. It may have been exposed to brute-force discovery bots, that go through the address space trying to find users.
Along the same logic: My car never got stolen. Not once. I always park on bright, accessible locations, near active streets. People who get their car stolen are guilty, 'cause they park in back-alleys near gangster groups.
Only that my car stereo was stolen at 4PM on a busy street on a working day last week...
Ask any sysadmin of a large email domain. You'll be surprised at the kind of attacks they see. And at the costs of bandwidth they pay because of spam. Bots aren't nearly the only way to send spam.
The way you say that code, released under the GPL, is not controlled even by its owner/creator, is misleading.
True, after releasing code under the GPL, the owner of the code can't 'take it back'. It is GPLed and can be used by third parties under the license terms.
However, the owner is free to license the original code under *any* other license -- namely under commercial licenses. The GPL does not exclude other licensing schemes.
One good case is the Mozilla code, which is licensed under a number of different licenses, where one of them -- the MPL -- allows them to create the proprietary Netscape browser.
Look. In 10 years of development, I've yet to see any window manager or desktop environment provide any sort of worthwhile user experience. All we have is software made by geeks for geeks.
That is *your* experience. I have been using Windowmaker happily for four years now. It mimics NEXT's interface, arguably one of the best user-interfaces out there.
Is it windows? No. Would a windows' user feel at home? No, of course not. That doesn't make it a geek environment, however.
We need *A* project to focus on. Get it right ONCE,
Please understand the "right" is different for different people. I personally hate the start-button/taskbar oriented interface of windows. Other people love it. I can't work without virtual desktops. Other people prefer having all apps at the reach of a mouse click...
Uniformization would be quite bad. And as I explained in a post above, it would unlikely produce results *that* much better.
You would be correct, weren't your logic based on a false preposition. You assume that focusing more effort in one project makes the project evolve linearly faster.
Team software development generates lots of sincronization points between team members. A large team loses more time communicating than developing. Thus, project development speed doesn't grow linearly with added manpower. This is one of the causes for Brooks's Law (the other being team setup-time).
So, how far from reality you are depends on how big a distributed software development team can get, before "thrashing". I'd say the dynamic nature of open-source projects already generates ideally sized teams for popular projects. There would be no advantage in making them bigger.
We need a single distribution. That's right. We need totally focused efforts.
No - we - don't
Competition is essential to pressure evolution. Even MS knows this, and promotes internal competition, to compensate for its monopoly status. Trying to mimic MS, however is not feasible. Linux doesn't have the slack MS's bank account provides. External competition is then the only viable option - and let the market filter out inefficient companies.
We need *standards* - for stuff that can be standardized. Filesystem hierarchies, file formats, etc.
Having dozens of interoperable distributions is really our best scenario, and linux is headed that way.
We need a single desktop.
Nope. We need a desktop standard API, for the basic stuff. Adding menu options et al. Forcing people to one desktop (directly or indirectly) is not an option. I though this was obvious...
I have ReiserFS on the server farm for a free webmail provider I manage (150kaccounts). It has proven very reliable, and fast enough that disks are not a bottleneck.
On my laptop, after reading the scary warnings on Gentoo install docs, I opted for XFS (not really, it was purely an excuse). I had a buggy intel e100 pro ethernet driver, and on many of the crashes, I lost all the data on some of the open files.
Filesystem integrity was ok, after every crash, but I never had the same data-loss behaviour with ReiserFS.
My advice is: benchmark the filesystems you consider stable, against your usage pattern. It's the only real data you need, apart from reliability info. However, if you're going to serve mostly static HTML, I'd say your bottleneck will be bandwidth or RAM, not your disks.
I was not very impressed with XFS, but it's my opinion only. It's credited as a Very Good Filesystem (tm).
Replace "Linux on the desktop" with Mozilla. Remember back to the M## era comments on slashdot. Now, realize that Mozilla got to 1.0, and is one hell of a browser. Magically predict that "Linux on the desktop" will come of age, albeit slowly.
OSS is not always written as fast as commercialware (although Mozilla was very fastly written). However, after gaining momentum, it is very difficult for a project to stop before it reaches the optimum feature set (like TeX and LaTeX).
A quick install of Nero Burning Rom, and I was able to make a backup copy of my game CDs. (I don't like taking originals to LANs where they can get destroyed or stolen).
Yeah, that's also my problem. I make copies of my friends' original CDs so they don't get scratched at my home.
- Yes, Officer, this is an original borrowed CD. It is a copy? Oh, yeah, I was afraid the original got scratched. Pirate? Who? Me? What do you mean, pirate?
Thank goodness Java doesn't have operator overloading. It's really one of those features that are there to make code writing 10 times easier, and code reading 100 times more difficult.
1+1... eval... 3???
Re:Installation vs. Usage - Mac 10 Windows 7, Linu
on
Coursey on Palladium
·
· Score: 2, Informative
Your oppinion reflects Linux as it was two years ago. Please do try and take both SuSE and Mandrake for a run. You'll see their TCP/IP dialogs don't mostly suck, and SuSE's admin interface is just plain excellent. Installation couldn't be easier (SuSE 8 default installs with three mouse clicks)
You use Debian, and I respect the choice. Its an excellent distribution, very easy to maintain *if* you're a techie. I've used SuSE on my desktop, and switched to Gentoo. I think its better *for me*. Debian and gentoo are both excelent. Neither one is advisable for my parents.
However, I have them using SuSE 7.3, with KDE 3. Again, not my choice of desktop or OS, but excellent for them. The machine never breaks as it used to with Windows, and they can do their work (mainly word processing, spreadsheets and digital image downloading). Linux is reaching readyness for the desktop. And it can only get better.
the difference in this case is that the ASP.NET code is calling the same drawing functions that any other non-web application would be calling.. not calling through a wrapper etc).
The big question now is:
How does a BSODed Webserver look like, when viewed from a browser?
Believe it or not, once at the London Tube, I experienced a reboot of a subway. When the train stopped in the station, the doors wouldn't open. The driver said on the loudspeaker he was trying to solve the problem, and 30s later, the train completely powered off (engines, lights, a/c). A few seconds, a power up, and voliá... Freedom again:-D
You are not comparing equivalent languages. C can't be compared to VB. A more adequate comparison would be VB and Python (perhaps with Qt or GTK bindings).
I don't see how Python is any more difficult to program than VB. It's just that stuff you don't know seems more complex that what you're used to.
You, or fellow developers, I am afraid to say, are the reason I have to wait *ages* at my local hypermarket to get the groceries checked out. But I guess you'll only get it when your app is in production, with 200 terminals. By then, it'll be too late.
Slashdot Mirror
I am a Palm III and Palm Vx owner. I carried a Palm for the last ~4 years. Its a good product, but one must realize that this is the computer industry. One doesn't sit tight on a good product for five years.
PPCs on the other hand, are horrendous devices. MS didn't resist on slapping the desktop GUI on those things. How dumber can you go over adding a taskbar on such a small screen?!
I believe PPCs are inferiour products. For equal evolution, Palm would give a much better platform (yes, I've programmed both PalmOS and PPCs). Monopoly has nothing to do with it.
Aw, gosh. I'm just feeding the troll...
Product development since then? Zero, zip, niente, nada de nada. They let go all of their competent techies, and are now a mass of marketeers without guindance, slowly sinking to the sound of Titanic's band.
It's really sad that these guys took Psion's market, and then managed to give it away to M$.
frodo:$ dig whitehouse.gov soa
;; res options: init recurs defnam dnsrch
;; got answer:
;; ->>HEADER ;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 0
;; QUERY SECTION:
;; whitehouse.gov, type = SOA, class = IN
;; ANSWER SECTION:
;; Total query time: 476 msec
;; FROM: frodo to SERVER: default -- 127.0.0.1
;; WHEN: Sat Sep 21 15:10:23 2002
;; MSG SIZE sent: 32 rcvd: 88
; > DiG 8.3 > whitehouse.gov soa
whitehouse.gov. 1D IN SOA eopc.eop.gov. postmaster.whitehouse.gov. (
2002072201 ; serial
15M ; refresh
5M ; retry
1W ; expiry
2H ) ; minimum
You know, if you hide the root servers DNS stops working, don't you?!
This would tipically be done on the data-link layer, but can also be done at the IP level.
If you can't change the software on the other end, you'd have to do the unification at the application layer. It depends on the application. Ex: for HTTP, you can balance the requests over the lines. When the size is known, you can do partial requests and balance these along the lines; for FTP, do PART requests for the pieces of the file. Expose a proxy for each application protocol. Not all applications can be adapted to take advantage of the multiplexing, however.
Portuguese laws forbid different entities from cross-referencing their databases, without explicit approval of the citizens. The way it is written, it even affects different departments of the state -- leading to a social security number, a tax ID number, etc.
I think it is a lot worse the way its done in the US, where everyone and their dog knows your SS#. It is very easy to cross-reference the DBs. At least here, they'd need to do some data mining...
His biggest error probably was talking about it. He should have sold the info to some mobster gang. They'd probably be much more gratefull.
Large email domains all have this problem.
Only that my car stereo was stolen at 4PM on a busy street on a working day last week...
Ask any sysadmin of a large email domain. You'll be surprised at the kind of attacks they see. And at the costs of bandwidth they pay because of spam. Bots aren't nearly the only way to send spam.
However, the owner is free to license the original code under *any* other license -- namely under commercial licenses. The GPL does not exclude other licensing schemes.
One good case is the Mozilla code, which is licensed under a number of different licenses, where one of them -- the MPL -- allows them to create the proprietary Netscape browser.
Is it windows? No. Would a windows' user feel at home? No, of course not. That doesn't make it a geek environment, however.
Please understand the "right" is different for different people. I personally hate the start-button/taskbar oriented interface of windows. Other people love it. I can't work without virtual desktops. Other people prefer having all apps at the reach of a mouse click...Uniformization would be quite bad. And as I explained in a post above, it would unlikely produce results *that* much better.
Team software development generates lots of sincronization points between team members. A large team loses more time communicating than developing. Thus, project development speed doesn't grow linearly with added manpower. This is one of the causes for Brooks's Law (the other being team setup-time).
So, how far from reality you are depends on how big a distributed software development team can get, before "thrashing". I'd say the dynamic nature of open-source projects already generates ideally sized teams for popular projects. There would be no advantage in making them bigger.
Competition is essential to pressure evolution. Even MS knows this, and promotes internal competition, to compensate for its monopoly status. Trying to mimic MS, however is not feasible. Linux doesn't have the slack MS's bank account provides. External competition is then the only viable option - and let the market filter out inefficient companies.
We need *standards* - for stuff that can be standardized. Filesystem hierarchies, file formats, etc.
Having dozens of interoperable distributions is really our best scenario, and linux is headed that way.
Nope. We need a desktop standard API, for the basic stuff. Adding menu options et al. Forcing people to one desktop (directly or indirectly) is not an option. I though this was obvious...On my laptop, after reading the scary warnings on Gentoo install docs, I opted for XFS (not really, it was purely an excuse). I had a buggy intel e100 pro ethernet driver, and on many of the crashes, I lost all the data on some of the open files. Filesystem integrity was ok, after every crash, but I never had the same data-loss behaviour with ReiserFS.
My advice is: benchmark the filesystems you consider stable, against your usage pattern. It's the only real data you need, apart from reliability info. However, if you're going to serve mostly static HTML, I'd say your bottleneck will be bandwidth or RAM, not your disks.
I was not very impressed with XFS, but it's my opinion only. It's credited as a Very Good Filesystem (tm).
OSS is not always written as fast as commercialware (although Mozilla was very fastly written). However, after gaining momentum, it is very difficult for a project to stop before it reaches the optimum feature set (like TeX and LaTeX).
- Yes, Officer, this is an original borrowed CD. It is a copy? Oh, yeah, I was afraid the original got scratched. Pirate? Who? Me? What do you mean, pirate?
*clunk*
1+1... eval ... 3???
You use Debian, and I respect the choice. Its an excellent distribution, very easy to maintain *if* you're a techie. I've used SuSE on my desktop, and switched to Gentoo. I think its better *for me*. Debian and gentoo are both excelent. Neither one is advisable for my parents.
However, I have them using SuSE 7.3, with KDE 3. Again, not my choice of desktop or OS, but excellent for them. The machine never breaks as it used to with Windows, and they can do their work (mainly word processing, spreadsheets and digital image downloading). Linux is reaching readyness for the desktop. And it can only get better.
In fact, it's what I'm using in my brand new 1.4Ghz laptop, so I guess it's also my choice for fast hardware.
The big question now is:
How does a BSODed Webserver look like, when viewed from a browser?
Believe it or not, once at the London Tube, I experienced a reboot of a subway. When the train stopped in the station, the doors wouldn't open. The driver said on the loudspeaker he was trying to solve the problem, and 30s later, the train completely powered off (engines, lights, a/c). A few seconds, a power up, and voliá... Freedom again :-D
I don't see how Python is any more difficult to program than VB. It's just that stuff you don't know seems more complex that what you're used to.
You, or fellow developers, I am afraid to say, are the reason I have to wait *ages* at my local hypermarket to get the groceries checked out. But I guess you'll only get it when your app is in production, with 200 terminals. By then, it'll be too late.
FP
This is an European mirror. If you're on the US, you are better off using ftp.suse.com directly or another mirror on that side of the atlantic.