The filesystem that BeOS uses is database oriented and allows complex queries in (basically) real time over the whole file system and it's metadata. I've used it in the past and was amazed.
This is the best link I could find so far that tries to explain the advantages:
In every case I've seen UML used for modeling, it has created systems which looked clean in the diagrams but which failed to function usefully once implemented due to lack of conceptual underpinnings.
On the other hand it's quite possible that the person who designed those systems wasn't very good at OO design and software development. A bad design is a bad design. It doesn't matter if you explain it using UML or plain text.
Your argument that UML is bad because every system you've seen documented using it was bad doesn't hold up.
a) not crash
b) alert the user, who presumably can do something about it
Putting up an alert box requires allocating memory, which you cannot do because you are
already out of memory.
Then free up some memory first! Chances are that you already allocated a heap of memory before failing. Now that you can't finish the operation, free the already allocated mem and then display the error message.
Basically cleanup and free any now unneeded resources and then display the error.
The one big difference is that if a company goes boom none of the software source is available for people to hack at and let live - it just disappears into the trash-heap of time.
Somehow you managed to miss the biggest and most important difference. Development of open source software depends on programmer/user 'mind-share' and not so much on profits from sales.
The NSA has the Win2k source code. It's very easy for universities and other establishments to get the source, slightly less easy for large companies, and slightly less easy still for small companies and individuals (although they're changing this as we speak...)
True, but just because they have the source doesn't mean they can hack on it or fork it like they can with Linux.
Okay, I see what you mean. Half points. The original poster wanted, I mean *needed* something that would allow arbitrary R/W ability to a given smartcard. What you have shown allows one to talk to a smartcard, and the programmable smartcards that come with those devices will let you read/write everything one them, but hey, guess what, the smart cards in your cell phone or pay TV won't let you R/W anything on them. Those $49 reader/writers won't let you get past that security restriction.
Remember, smart cards are programmable. They can be programmed to not talk to strangers bascially...
Adding this kind of copy protection to PCs, set top, boxes etc is basically the same at the music industry's attempts at a secure digital format to replace CDs. You can't add take an existing product (i.e. CDs or PCs), add copy protection and expect it to succeed in the marketplace against it's un-secured cousins. Secured versions of existing tech offer no extra value for consumers. It's that simple.
The only way to get copy protection into people's homes is to piggy back it on something new and cool that people might actually want. DVD being a good example. High quality movies, ~7 Gig disks. Something that wasn't possible or available before. People want that, can't get it elsewhere, and will put up with CSS and the other annoying copy protections features that come with it.
There is no such thing as a "cheap smart card reader/writer". You are probably thinking of magnetic strip cards which have been around for ages and definately are *not* smart. Smart cards basically are tiny computers on a chip. You can talk to the smart card and communicate with it, but you certainly can't just read/write it's memory. In fact you have no direct access to it's memory at all. As someone else here mentioned, Smart cards are designed so that if you try to tamper with them (physically or whatever) thier memory will be erased. Program, memory contents, and encryption keys, gone...
I just checked out my dopey, USENET posting past, and I must say that it's a very weird feeling once again seeing messages and discussions that I thought were long dead.
Also remember that people are also archiving large chunks of the web purely to capture a piece of "history".
This was on the securityfocus incidents list the other week.
In the past two months somewhere.com has received over 300,000
misdirected mail messages.
In the past 24 hours my top ten is
786 someone@somewhere.com
107 name@somewhere.com
85 somebody@somewhere.com
78 me@somewhere.com
78 nowhere@somewhere.com
70 bounced@somewhere.com
65 kelly@somewhere.com
63 somone@somewhere.com
61 PianoMan52357298@somewhere.com
50 something@somewhere.com
There is a lot of good stuff in this interview covering many areas. This is IMHO a really interesting point that no one has mentioned yet. It explains why Linux and Free software in general maybe one of the most important things to happen to poorer nations trying to moving into the "Information Age", and also why they might end up kicking our butts in the West...
MOGLEN: Well, if you look at the computer science 101 syllabi of universities in India, Southeast Asia, Indonesia, if you go on the web and look at these, the CS 101 curriculum in these universities assumes that people are using a Linux based computer rather than a Windows based computer. I mean, it's free. So, Singapore and Thailand and Malaysia are going to produce a lot of young adults who learned about computers using free software; the computers in their homes are going to be free-software computers; their children are going to grow up with free software computers. Which bunch of people are going to be the talented, engaging, aggressive programmers, busy making changes?
WORTHINGTON: So, culturally, you see the collective decision to use Windows as one which forecloses the possibilities of generations down the road?
MOGLEN: Well, it's a decision to have fewer programmers. The whole point of free is freedom to change, not low cost, and the whole point of the world towards which we are moving is that the primary power distinction, the class line, is between those people who know how to change the behavior of computers and those people who don't. Because that kind of knowledge, in particular, the ability to interact with complex technological systems to alter their behavior, is power over ordinary daily life in a profound way.
The DOS history is also missing all the other things that went into DOS/Windows over time. Like chunks of 'Stacker' and all the other 3rd party products that were bought out and assimiliated.
How many other products that went into Windows can you name?
--
Simon
Re:This is berlin without the complications?
on
XFree & Rendering
·
· Score: 1
I disagree. Since no-one in thier right mind codes directly on top of X and uses a toolkit like GTK or Qt, getting a large (and growing) amount of software over to Berlin would mostly be a matter of porting GTK and Qt to Berlin and then recompiling.
> Take the "mount manager" tool. How is this tool
> different than a text-editor with some shortcuts
> to the applicable files (/etc/fstab,
>/etc/smb.conf, etc)? Answer: It is only easier
> to use to someone who has never seen it before
> but it is less powerful AND not forwards
> compatible.
I would argue that it is more powerful precisely
*because* it is easier to use for someone who has
never seen it before.
> Only Easier The First Time: The GUI looks
> EXACTLY the same with the exception that the
> options are all laid out as checkboxes. But the
> options aren't explained, so no power is really
> gained here.
Like smeg no power is gained! The power is that I
don't have to know some bullshit syntax and I
can just click on what I want an not even risk
screwing up the syntax of the whole config file.
> How does it help to know I have a
> "password" option for/dev/hda1 if I don't know
> what that means?
F1 for docs? those 'mouse tip' thingies. This is worse than a text file in what way?
> Less Power: The tool doesn't give you any
> information how the file is actually laid out
> and there is no integrated text-processing.
> So the user does not advance on any path of
> knowledge where they can become MORE efficient
> (through the use of scripts, etc).
Why the hell should I care about the layout of the
config file?!?! I shouldn't have to know how to
write a script to hack the config in order to just
get the bloody thing to do to do anything useful.
> Not Forwards Compatible: If a new filesystem
> came along that had extra options, this tool
> would have to be re-written to accomodate.
> Whereas/etc/fstab wouldn't have to change one
> whit.
So what? The benefits of GUI clearly out way the disadvantages.
Any 'tool' wizz kid puts together a web page and sticks it up on the net and world thinks he's a bloody computer programmer genius super-hacker. WTF!?!
When I was thier age (I'm 24 now), I taught myself C and Asm (68k) by *myself*. And I can tell you now that there was anyone there that I could ask for help, and it sure as hell wasn't trendy either.
What's changed? Well the internet is here and it gives a woody to any suit who has a bit of money to invest and wants to make a quick buck. So any little smartass that do anything with web is treated like some sort of smegging god.
Programming HTML? bah humbug. Learn ya self a real language kid.
(Powers That Be, can we please have a survey of programming languages learnt, age, and whether they were learnt in school, at home, by oneself etc)
(Disclaimer:I'm no expert on jamming radio communications.) Is it difficult to jam the mobile phone system in a local area? I imagine all you would have to do is disrupt communications between the phone and the tower such that the phone doesn't manage to get on the network. (In much that same way that you can kill arbitrary TCP connections on ethernet by injecting FINs) Hell, you wouldn't even have to transmit all the time. You could just detect a phone searching for a tower and jam it then to stop connection.
The best part about this system is that it would be backwards compatible with older phones, doesn't require getting manufactures on board and losers can't override it.
I got it to work ok. I just followed the Redhat install via RPM instructions while not installing the 'menu' RPM stuff. Here is a list of what I installed and the order I used:
Gimp has been in feature freeze for the last 3 or so developer versions. As well as being beta tested and heavily debugged. Gimp 1.2 will be here Real Soon Now.
BTW, is it just me, or does PhotoPaint run slower than a slug?
I don't think that apps are quite as important as they might first appear. GTK/Gnome apps seem to have no problem running under KDE and visa versa. People can choose whatever desktop they like and still run whatever apps they like. Isn't that the kind of freedom and choice that Linux/Free software/Open Source is about?
Also from what I can tell, the KDE and Gnome developers both recognise the importance of interoperability between KDE and Gnome stuff and are working towards this goal. Not too mention the fact that (for the most part) KDE and Gnome are both under GPL so people can freely port apps back and forth between the two.
There is no reason for everyone to converge on only one desktop. Both can live happily.
not quite. The point was that just because someone knows how to break a bank vault doesn't mean they know how to make an unbreakable vault. You don't hire Billy the Kid to make a better bank vault. But you *do* hire him to work in the QA department of the place that does make vaults.;-)
BTW, maynard, Harris Miller did *not* recommend a central email cleaning thingy. He meantioned it, but did not recommend it. Mind you he didn't reject the idea on privacy grounds either..
I spend almost all my time at work developing sites using ASP, PHP, sometimes Perl etc. So I'll try to be fair here with some comparisions.
(Disclaimer: I haven't used the most recent version of ASP, so my comments may be out of date).
I theory I don't mind ASP too much. By using Javascript (JScript in MS speak) instead VBScript you can avoid any real BASIC brain damage risk, and also you can use OOP which becomes invaluable for larger projects.
ASP:
* Feature starved. You can't even do file upload without 3rd party software. ASP doesn't come with a big range of libraries to use like PHP does. Although you can use all that COM, ActiveX stuff, but that often requires having to use 3rd party software. (Read: extra expense, and you can't be sure that the Web host you deploy on actually has the extra software you need installed). Simply put, you don't have everything you need right in front of you like you do with PHP.
* Buggy. At least the JScript stuff is. String objects often forget half thier methods. Oh, and don't forget to copy the objects out a DB ResultSet otherwise you are in for some very weird bugs later on.
* Documentation. At least for the JScript part it is a complete work of fiction.
* Portablity. You are basically stuck with NT. Unix versions are around but they cost a fair bit and I would be worried about compatability between different implentations. (Actually I've had trouble between two machines with I *thought* were running the same version of IIS and ASP).
* Cost. You need NT to run it.
The bottom line is that PHP is a much better implementation of the approach shown in ASP. It is very stable, easy to use C like language, has fully featured libraries for things from Mail, regex, DB, XML, IMAP you name it, good documentation (check out the Annotated Manual at ww.php.net), is portable, available at the right price and is under an Open Source liscense. It's just much better.
I have to agree, I didn't expect to see come from Metallica either...
Time for a technical question. Does anyone know how 'fingering' a user on Napster works? Can it be stopped/detected at the server. If so, then Napster could then enforce thier "no bots on the service" rule. Right back atcha'!
Slashdot Mirror
The filesystem that BeOS uses is database oriented and allows complex queries in (basically) real time over the whole file system and it's metadata. I've used it in the past and was amazed.
e =1 3
This is the best link I could find so far that tries to explain the advantages:
http://www.osnews.com/story.php?news_id=421&pag
I just hope that Linuxland listens to Reiser sooner instead of later. He does have the vision to bring some real progress to filesystems.
--
Simon
"every system I've seen documented in UML was bad" therefore "UML is bad".
That is the reasoning I believe is faulty. Clear?
--
Simon
On the other hand it's quite possible that the person who designed those systems wasn't very good at OO design and software development. A bad design is a bad design. It doesn't matter if you explain it using UML or plain text.
Your argument that UML is bad because every system you've seen documented using it was bad doesn't hold up.
--
Simon
Basically cleanup and free any now unneeded resources and then display the error.
Don't be so bloody helpness. sheeessh...
--
Simon
Somehow you managed to miss the biggest and most important difference. Development of open source software depends on programmer/user 'mind-share' and not so much on profits from sales.
--
Simon
True, but just because they have the source doesn't mean they can hack on it or fork it like they can with Linux.
--
Simon
Okay, I see what you mean. Half points. The original poster wanted, I mean *needed* something that would allow arbitrary R/W ability to a given smartcard. What you have shown allows one to talk to a smartcard, and the programmable smartcards that come with those devices will let you read/write everything one them, but hey, guess what, the smart cards in your cell phone or pay TV won't let you R/W anything on them. Those $49 reader/writers won't let you get past that security restriction.
Remember, smart cards are programmable. They can be programmed to not talk to strangers bascially...
--
Simon
Adding this kind of copy protection to PCs, set top, boxes etc is basically the same at the music industry's attempts at a secure digital format to replace CDs. You can't add take an existing product (i.e. CDs or PCs), add copy protection and expect it to succeed in the marketplace against it's un-secured cousins. Secured versions of existing tech offer no extra value for consumers. It's that simple.
The only way to get copy protection into people's homes is to piggy back it on something new and cool that people might actually want. DVD being a good example. High quality movies, ~7 Gig disks. Something that wasn't possible or available before. People want that, can't get it elsewhere, and will put up with CSS and the other annoying copy protections features that come with it.
--
Simon
There is no such thing as a "cheap smart card reader/writer". You are probably thinking of magnetic strip cards which have been around for ages and definately are *not* smart. Smart cards basically are tiny computers on a chip. You can talk to the smart card and communicate with it, but you certainly can't just read/write it's memory. In fact you have no direct access to it's memory at all. As someone else here mentioned, Smart cards are designed so that if you try to tamper with them (physically or whatever) thier memory will be erased. Program, memory contents, and encryption keys, gone...
No one has found a practical attack yet. AFAIK.
--
Simon
I just checked out my dopey, USENET posting past, and I must say that it's a very weird feeling once again seeing messages and discussions that I thought were long dead.
Also remember that people are also archiving large chunks of the web purely to capture a piece of "history".
--
Simon
--
Simon
--
Simon
The DOS history is also missing all the other things that went into DOS/Windows over time. Like chunks of 'Stacker' and all the other 3rd party products that were bought out and assimiliated.
How many other products that went into Windows can you name?
--
Simon
I disagree. Since no-one in thier right mind codes directly on top of X and uses a toolkit like GTK or Qt, getting a large (and growing) amount of software over to Berlin would mostly be a matter of porting GTK and Qt to Berlin and then recompiling.
Yes thanks,
--
Simon
> Take the "mount manager" tool. How is this tool /etc/smb.conf, etc)? Answer: It is only easier
/dev/hda1 if I don't know
/etc/fstab wouldn't have to change one
> different than a text-editor with some shortcuts
> to the applicable files (/etc/fstab,
>
> to use to someone who has never seen it before
> but it is less powerful AND not forwards
> compatible.
I would argue that it is more powerful precisely
*because* it is easier to use for someone who has
never seen it before.
> Only Easier The First Time: The GUI looks
> EXACTLY the same with the exception that the
> options are all laid out as checkboxes. But the
> options aren't explained, so no power is really
> gained here.
Like smeg no power is gained! The power is that I
don't have to know some bullshit syntax and I
can just click on what I want an not even risk
screwing up the syntax of the whole config file.
> How does it help to know I have a
> "password" option for
> what that means?
F1 for docs? those 'mouse tip' thingies. This is worse than a text file in what way?
> Less Power: The tool doesn't give you any
> information how the file is actually laid out
> and there is no integrated text-processing.
> So the user does not advance on any path of
> knowledge where they can become MORE efficient
> (through the use of scripts, etc).
Why the hell should I care about the layout of the
config file?!?! I shouldn't have to know how to
write a script to hack the config in order to just
get the bloody thing to do to do anything useful.
> Not Forwards Compatible: If a new filesystem
> came along that had extra options, this tool
> would have to be re-written to accomodate.
> Whereas
> whit.
So what? The benefits of GUI clearly out way the disadvantages.
--
Simon
Any 'tool' wizz kid puts together a web page and sticks it up on the net and world thinks he's a bloody computer programmer genius super-hacker. WTF!?!
When I was thier age (I'm 24 now), I taught myself C and Asm (68k) by *myself*. And I can tell you now that there was anyone there that I could ask for help, and it sure as hell wasn't trendy either.
What's changed? Well the internet is here and it gives a woody to any suit who has a bit of money to invest and wants to make a quick buck. So any little smartass that do anything with web is treated like some sort of smegging god.
Programming HTML? bah humbug. Learn ya self a real language kid.
(Powers That Be, can we please have a survey of programming languages learnt, age, and whether they were learnt in school, at home, by oneself etc)
(Just venting a little bit.)
--
Simon
(Disclaimer:I'm no expert on jamming radio communications.) Is it difficult to jam the mobile phone system in a local area? I imagine all you would have to do is disrupt communications between the phone and the tower such that the phone doesn't manage to get on the network. (In much that same way that you can kill arbitrary TCP connections on ethernet by injecting FINs) Hell, you wouldn't even have to transmit all the time. You could just detect a phone searching for a tower and jam it then to stop connection.
The best part about this system is that it would be backwards compatible with older phones, doesn't require getting manufactures on board and losers can't override it.
>;-)
--
Simon
I got it to work ok. I just followed the Redhat install via RPM instructions while not installing the 'menu' RPM stuff. Here is a list of what I installed and the order I used:
libwine-graphics9-glibc-2.1
wine-graphics9-glibc-2.1
wpo2000-fonts-core
fonttastic-glibc-2.1
libaps
graphics9-paint
graphics9-common
graphics9-help-paint
graphics9-help-techsupport
graphics9-help-common
I ran it from the K menu too.
--
Simon
Gimp has been in feature freeze for the last 3 or so developer versions. As well as being beta tested and heavily debugged. Gimp 1.2 will be here Real Soon Now.
BTW, is it just me, or does PhotoPaint run slower than a slug?
--
Simon
...you forgot "loses laptops holding classified nuclear weapons data."
Have a nice day.
--
Simon.
I don't think that apps are quite as important as they might first appear. GTK/Gnome apps seem to have no problem running under KDE and visa versa. People can choose whatever desktop they like and still run whatever apps they like. Isn't that the kind of freedom and choice that Linux/Free software/Open Source is about?
Also from what I can tell, the KDE and Gnome developers both recognise the importance of interoperability between KDE and Gnome stuff and are working towards this goal. Not too mention the fact that (for the most part) KDE and Gnome are both under GPL so people can freely port apps back and forth between the two.
There is no reason for everyone to converge on only one desktop. Both can live happily.
--
Simon
not quite. The point was that just because someone knows how to break a bank vault doesn't mean they know how to make an unbreakable vault. You don't hire Billy the Kid to make a better bank vault. But you *do* hire him to work in the QA department of the place that does make vaults. ;-)
BTW, maynard, Harris Miller did *not* recommend a central email cleaning thingy. He meantioned it, but did not recommend it. Mind you he didn't reject the idea on privacy grounds either..
--
Simon
PostgreSQL is under an Open Source license too. So the cost is the same. Now why would I then choose MySQL over PostgreSQL?
I think it's more a case of people not yet clue-ing up to the fact that PostgreSQL is actually the best free DB at the moment.
--
Simon
I spend almost all my time at work developing sites using ASP, PHP, sometimes Perl etc. So I'll try to be fair here with some comparisions.
(Disclaimer: I haven't used the most recent version of ASP, so my comments may be out of date).
I theory I don't mind ASP too much. By using Javascript (JScript in MS speak) instead VBScript you can avoid any real BASIC brain damage risk, and also you can use OOP which becomes invaluable for larger projects.
ASP:
* Feature starved. You can't even do file upload without 3rd party software. ASP doesn't come with a big range of libraries to use like PHP does. Although you can use all that COM, ActiveX stuff, but that often requires having to use 3rd party software. (Read: extra expense, and you can't be sure that the Web host you deploy on actually has the extra software you need installed). Simply put, you don't have everything you need right in front of you like you do with PHP.
* Buggy. At least the JScript stuff is. String objects often forget half thier methods. Oh, and don't forget to copy the objects out a DB ResultSet otherwise you are in for some very weird bugs later on.
* Documentation. At least for the JScript part it is a complete work of fiction.
* Portablity. You are basically stuck with NT. Unix versions are around but they cost a fair bit and I would be worried about compatability between different implentations. (Actually I've had trouble between two machines with I *thought* were running the same version of IIS and ASP).
* Cost. You need NT to run it.
The bottom line is that PHP is a much better implementation of the approach shown in ASP. It is very stable, easy to use C like language, has fully featured libraries for things from Mail, regex, DB, XML, IMAP you name it, good documentation (check out the Annotated Manual at ww.php.net), is portable, available at the right price and is under an Open Source liscense. It's just much better.
--
Simon
I have to agree, I didn't expect to see come from Metallica either...
Time for a technical question. Does anyone know how 'fingering' a user on Napster works? Can it be stopped/detected at the server. If so, then Napster could then enforce thier "no bots on the service" rule. Right back atcha'!
--
Simon