That would be a law requireing some software to be used inside the goverment itself. The goverment of normal country cannot dictate businesses what software to use. And if it can - then there is somethung wrong with that country and that goverment in general, more generally then just a software or any technology at all. That place, where it's wrong, is called constitution. Go and fix THAT at first, than come back to decide about technologies.
As for the goverment choice for KDE against GNOME - I have no problem with it (again if I have my own choice). I don't like personally KDE, but I have no problem if the goverment will use it.
99% of typical Microsoft users do not know any difference between Linux and Open Source. And they don't know what are Open Standards. You can spend one more centure educating them that linux is just a kernel, there are many distros on a top of that kerne and there are other kernels as well. Or you can spend some more reasonable time advocating Linux as a concept keeping other Open Source" in mind. What do you prefer?
By the way, I remember such a comment from one of decision makers, when I said that some of our server may run BSD: "BSD? It's a sort of Linux, right? Well, let's decide about Linux in general and later we will specify which distro to use on which computer."
I know, some BSD funs may dislike it. But in many cases it's really easier to approve "Linux in general" and later specify that sometime it can be BSD:)
Well, historically Linux was stealing a lot open source applications from BSD world. And that time for many managers Linux was one of BSD kinds. Now time is changed.
Re:Lobby for Open Standards
on
Lobbying For Linux
·
· Score: 4, Insightful
It's easier to lobby Linux at first. It opens a mind of users. After that it's easier for them to notice other open source systems.
Besides, 100% of Linux is not a monoculture. There are dozens of distros. There is a competition between Gnome vs KDE vs misc managers. There are various programming languages to code the same projects. Well, even inside Perl there are always more than one way of doing the ame thing (according to Laryy Wall, the creator of Perl). Finally, there will be always PostgreSQL vs MySQL and Emacs vs vi (and even GNU/Emacs vs Xemacs for Emacs winners) - what kind of monoculture do you see here?
By the way, all those "one vs another" things are OS agnostic. If the decision maker will open the mind enough to get into Linux - in no time the person will notice that all those "vs" are the same inside BSD.
But if you will bring "Linux vs BSD" right to the lobbying process, those non-techs will decide: "OK, let's them at first decide Linux or BSD and THEN we'll see if it's worthy against Windows!" - you don't want THAT decision, don't you?
How many computers with Windows are you using along with your team? How about 50 or 100? That's my case, and my current employer is not the first place where the situation is the same. Well, some of them really never crash, but most of them are crashing at least on a weekly base, some evn on a daily base. The list of version being craching often: NT 4, W2K, XP, all in both workstation and server flavors.
My observation shows that the most stable ones usually are those where we rarely install anything besides a single application (or suit). If either Office or Photoshop or Visual Studio or MS SQL Server is the only one which is installed - than chances of crashing are bellow the average. Two different applications and it's a big chance that we've got a problem. That makes a correction about Windows being an OS for a personal computer - it for one personal application. I think that Microsoft's QA is testing it's operating system mostly installing one of MS applications. Bad QA.
Another source of problems we are usualy having on computers with 3rd party hardware drivers. You can blame 3rd party vendors, but why I don't have such problems with the same hardware on Linux? Oh, I know why: b/c with Linux the vendor has free (often) support of developers and QA from around the world. In case of Windows the vendor is on its own doing QA often without any support from Microsoft.
We can also talk more about:
NTFS - old FS, too bad nowaday comparing to Ext3, XFS, JFS and ReiserFS. What's wrong to get a modern FS if its more stable?
why can't I make a page file in the raw partiotion?
registry database is defenitely not designed for situations when you install/uninstall/reinstall many applications often. Your system will die or become slow soon if your job is installation/uninstallation/reinstallation of new software.
Windows updates are supposed to improve my system, not to crash it. But even after failing, deinstallation of the recently applied update can crash the system as well. And nothing you can do about it.
Windows is still using Netbios in LAN. Bad protocol, bad implementation. Wins server is too bad, especially in cache updates. Lmhosts must die already by now. Why not using DNS inside the LAN?
Why the whole system must be locked if Explorer is trying to connect to the slow LAN server or a remote web folder?
Eventlog very rarely gives any useful information. There should be a better control over debugging levels. Otherwise you feel yourself blind when it's crashing.
Don't tell me that I should not tinker my OS. tell it to ma&pa - home users. In a corporate environment you've got a documentation from 3rd party vendors requiring tinkering. And same level of tinkering doesn't crash the OS if ti's designed for that - I mean Linux.
Meme? I think that it's meme only when you do everything only if Microsoft said you to do. But the real life is not what Microsoft was planning. That's why people more and more often prefer the OS that is designed for being owned by net/sys-admins, not just licensed. By being owned I mean being controlled as required.
In my file of iptables rules 20% rules are related to NAT and forwarding, while 80% are related to in-out access in general. I've got my first rule-set from one of examples from the netfilter documentation, applying some specific changes, but even in the original file 80% of rules were about access in general, not NAT/forwarding specific.
Also, you can check the configuration of netfilter in your Linux kernel, or even the netfilter source code: less than 20% will be about NAT.
So, where did you get your knowledge about "80% of a firewall is NAT?"
1. The original topic was about PHP. That's why I originally have explained why to go from PHP to Zope. I did not compared Zope to Krisalis as today I first time heard about it (but I promise to read more about it some day).
2. If I would go with pure XSL solution that it will be Cocoon - many developers know it so it would be no problem to make a team. Cocoon has a very stable support from one of open source leaders: Apache. And Cocoon has the best XSL-oriented design for being used in content-processin applications. However, I don't think that "pure XSL" solution is the way to go. XSL doesn't cover everything. It's good for dynamic content processing, while typically I need a content management. How kristalis solves that problem?
3. The most important part missed in XSL is that functionality (as it's applied to a content) should be managed as a part of the content in a same dynamic way. That's why in Zope Python is a very essential part of DTML and ZPT. I know that some functionality can be implemented in XSL, but people who have some experience of building a very complicated systems with XSL can confirm: it's a pain in the neck. How is this issue addressed in Kristalis?
4. XSL is not designed as a general programming languages. But all attempts to extend it with non-scripting languages will almost always fail. You need a flexibility of a scripting language to embed it to tags on the fly. At the same time Perl and Tcl are not OOP enough, nor they are expressive/readable. That's why the choice is coming to Python in Zope. Can you compare it in details to Kristalis?
If you have to fix some existing PHP application - go for it. But if you have a chance to start the webapp over then do your best to make sure that your choice of Zope will be approved. Well, it's easier every day as Zope is gaining more users and success stories.
This is the excelent example of a comment from people who use OSX not because OSX is good (which is not bad), but because such people hate Windows.
I use all three desktops, Windows, (various flavors, including XP), Mac OS (including 9 and X) and Xfree (including GNOME, KDE as well as simple windos managers).
Standard disclaimer: IANADP (I am not a desktop programmer) - I don't do it for living, so my mind is not corrupted by desktop programming. Although, I still do it rarely and also although, I do software development for database, network and system management, for data and knowledge processing and for web-based content management, so I sue desktop as a end-user a lot (I mean really a lot). For example, I cut-n-paste not only text, but alose images and vector diagrams.
What I have noticed: all desktops have their own good and bad features. OSX reminds me Windows 3.1 as it's same unstable, but in a very modern themed. XP is another attempt of stylishing, although the real value of XP is in a combination of much longer debugged history (Windows today is much more stable) and still old bad misfeature of linking desktop parts to the kernel (Explorer, IE and CIFS are parts of desktop environment, right?). XFree desktops are the most stable unless you use OpenGL - a simple xscreenserver in its 3D theme can lock all your PC so much CPU it uses when something wrong. For end-users all three are the same.
Well, all said above about stability of desktops is not said about the underlying OS. Linux and BSD are the most stable OSes. OSX is the second (when it's not about GUI), Windows is the worst (we all know why).
Also, the stability of your work is a function of your applications. Specifically, how they are good with OS and do they still deliver features enough for your job on that OS. OpenOffice under Linux is the most stable Office application with still good enough feature set to my job.
Conclusion: if you really care about your job to be done than most likely you will never criticize some OS in a zealotish manner - instead you will always compare good and bad sides of both (or three) OSes.
But if you're really dying, try this: http://pear.php.net/python: "This extension allows the Python interpreter to be embedded inside of PHP, allowing for the instantiate and manipulation of Python objects from within PHP."
What's the point: the underlying PHP language is one of problems of PHP, but not the ony one. I can list them here with the solution I propose (Zope):
Language: sure Python is better choice for a readable code and OOP. In my personal opinion, Python is the best among imperative languages. I love using FP style in Python, getting the smaller code doing more and still very expressively.
Templates: Even DTML in Zope beats PHP. As for ZPT - after learning it you cannot go back to PHP. Never.
Namespaces: ZPT has a very excelent design and implementation of namespaces. All accordingly to W3C standards and traditions.
Content management: in PHP you still have to write it (or find it, or buy it). With Zope you have already got it, working, convinient, secure and well designed. CMF with Plone improve it even more bringing in workflow management and collaboration.
Database persistence: even if you don't care you already have it ZOPE. ZODB is object, transactional, reliable DBMS. If you care to integrate it with you existing SQL applications than you can use Adaptive Storage (APE) which gives you a transparent way to do it, similar to VFS+fstab give you filesystems in Linux.
I encourage disappointed PHP people to learn Zope. Even if your clueless boss will dictate you the choice of Sun MicroSCOftsystems, the value of concepts in Zope you learn cannot be overestimated - you can reuse patterns from Zope in all other other technologies, like ASP.
PHP code does not see (reflect) itself as data, no reflection in PHP. The code FUNCTIONS are not first class objects. So, that's why in PHP the code is not a data from FP (Functional Programming) prospective.
IN FP languages, like Lisp, the code is the data: there is very well defined reflection, I can construct new functions and manipulate existing ones as they are first class objects. Same/similar situation is in ML and Haskell.
Well, among traditionally imperative programming languages there are more and more cases of "the code is the data" paradigm as well. First of all in iterpreting (scripting) languages, like Python, Perl and Tcl. You can construct the text of the function and "eval()" it. The problem is that in FP languages there is a very well designed math model for it protecting you from many errors (you construct real functions, not a text for for functions). In scripting languages it's more like a hack leading to many errors in a similar way as C pointers and Fortran GOTO operators.
All three, PostgreSQL, MySQL and ZODB, are as portable as OpenOffice: they ARE already working on the same platforms as OpenOffice.
As for the part of you comment where you tells that BDB is more user friendly than PostgreSQL, I think you are trolling here. Unless you want to explain yourself.
The Adabas database application is not soemthing which can be replaced by BerekelyDB. They are very different things.
Why BDB? There is PostgreSQL if you love BSD license. And there is MySQL if you prefer GPL. And there is ZODB if you need OO layer on a top of BDB.
However, it is possible that Sun could write a whole new database application using BDB as the backend; or, Sun could write a layer for storing word processing, spreadsheet and other kinds of documents in BDB, affording some cool features that we don't get with flat binary files (which suck).
What kind of idiot would decide to do so? Only one who doesn't know (or doesn't want to know) about even one database I have mentioned above.
There are still many idiots in US thinking about a nuclear world war. Most of them have voted for the current president. Or another way around: most of current president votes were from such idiots.
As it is now EU is not capable to begin any serious wars. Not from military capability prospective - such decision would be politically impossible in EU. China is also not that stupid to through nuclear warheads here and there - they realize that that would be the end for all of us.
The problem is that US administration is driven by corporations, currently by those who is benefiting from any military race. And there is no way to stop them.
Tell me about it. But they are not just expensive - they are completely unreliable. In my prev/empl company they have bought about 50 Sun computers, from Ultra-10 to E450 to E4500. 80% of them had a hardware failure at least once within two years. Not HDD, but motherboard or one of controllers. As fo HDD - there is something seriusly wrong with Sun disk arrays - two many failurs, even with IBM HDDs which work much more stable in non-Sun arrays.
I asked my friends in other companies and people in mail-lists - the situation is always the same. And it's sad to compare IBM RS6K and HPPA equipment - people working with it are surpised hearing our statistics and complains.
It doesn't have to be THAT expensive if it's not more reliable than Intel same-performance servers. Or it doesn't have to be THAT unreliable for THAT price ans speed. Hmm, i am not sure if I need faster, but still expensive and THAT unreliable servers.
I guess someone should find the comparison between metro areas (no countries) to show you that NYC and LA are still worse than a dozen of other comparably same size cities. Of course per capita - that what counts by regular people comparing politicians.
When I buy the car, new or old, I have to get the license for it. The part of the procedure is to get the car checked for basic safity and environmental conditions. I the car makes a lot of noise or pollution or its brakes do not work - it cannot be licensed and i have to fix it (paying to the mechanics or doing it by myself).
Besides the car I have to get a driver license and I have to renew it from time to time. If I get too many penalty points or if I am noticed in one-time serious traffic violation my driving license can be suspended.
Same thing should be for comupters (AND networks) at the moment of connecting them to ISP:
each computer (or a whole network) must have a license to be connected to Internet;
periodically (as often as it's appropriate) safity and environmental checks must be take care:
the computer must be protected in terms of ports opened and mail filters installed;
the nightly based cron procedure must do the check and alert ISP if anything wrong found;
from time to time (weekly or so) ISP must scan clients from outside;
Besides my computer, my Internet and PC end-user skills must be licensed:
if I don't know how to update my OS or to install a security patch on it then I cannot be licensed;
if my PC is noticed as a source of virus/attacks and it's proven it's been cracked/infected than I've got my penalty points and have to pay a small fine (bellow a hundred of $);
if I am noticed as distributing viruses knowingly or hacking myself then my license should be suspended - I have to get the end-user class again, renew my license and pay a big fine (thousands of $);
IMHO it will improve overall Internet safity (imagine how much less there will be opened port and unpatched computers) and accelerate the whole national economy (imaging how many "mechanics garage" companies will rise their revenue!). It will open many new IT jobs and improve exisiting ones (now my boss cannot tell me "I don't belief some one may crack us - we don't have any useful information"). By the end of the day companies will actually safe money as they spend too much now to fight security in such insecure world.
Also it will improve the competition on the market as people will prefer more secure OS to be installed on their PCs. Oops, Bill gates may hate it and lobyy against it. But I still love the idea.
You will be charged for your harddrive as well proportionally to its size AND to the statistical chance how many time per HDD life the diskspace has been reused by removing old MP3 files.
For 70GB HDD, which is essentially equal to 100 CDs, and the statistical chance that the space has been reused 100 times per HDD lifetime (quite reasonable for let's say 3 years) you will be charged 10,000 times of what you would be charged per CD.
Not bad.
Re:Lisp and Python are compilers AND interpreters
on
Secure Programming
·
· Score: 1
You can have compiled, interactive implementations (some Common Lisp implementations), and you can have interpreted, non-interactive implementations (your typical Java VM).
Interesting. Any links or your own words explaining it in deeper details? Thank you in advance.
Lisp and Python are compilers AND interpreters
on
Secure Programming
·
· Score: 1
I am reminding from time to time also:
Lisp and Python has both byte-compilation and interactive-loop interpretation. If you haven't tried yet the interpretation mode in Lisp that doesn't mean it doesn't exist in Lisp. It's even described in most of Lisp books I've seen. It's very convinient when you need a sort of interractive debug mode.
I agree that Python byte-compiled code is still not as fast as C. However, in most of application (if they are not mission critical kernel drivers or DB engines) you don't really need to squeeze that performance. But I would not say that Python bytecode is always THAT slower than Lisp bytecode.
Re:We really need a different language
on
Secure Programming
·
· Score: 1
Python is good for some things, but it's hardly a 1:1 replacement for C/C++. While it may be somewhat better in some areas, interpreted languages have certain shortcomings that make them unsuitable as replacements for compiled languages in many cases. Such as anything rquiring small size or fast execution...
I think you exagurate a lot here, when you use the word many to advocate C, and the word some applying to Python and other interpreters. I think that in most of projects (by amount of projects and by amount of programmers assigned to) good interpreters (like Python or Lisp or Erlang) are more preferable as they let coding in safity, they let coding fast and they cut the cost of further maintainance.
Tasks that rquiring small size or fast execution are very rarely cases, such as kernel drivers and DB engines. In most applications C is overkill in terms of compactness and speed. In some programs C is the best choice.
Modern interpretors are also more effective than before. For example, check Zope content management and application server and its companion ZODB object database. Both are written in Python. Zope when running is comparable by size to full-loaded Apache (vs typical Java based application servers), while by speed it's not (much) slower than JBoss.
Same good things can be said also about interpreters of Lisp and Erlang, another full-featured generic programming languages used in many commercial applications.
If Zope is open-source, why not port the Zope framework over to PHP?
Very interesting logic: every open source project must be implemented in PHP.
Zope is a seriously designed application server. The major concept of Zope is to separate aspects. OOP is important concept but it is not the concept theat the programmer is forced to use exclusively. Also, in Zope the stateful and stateless content is managed dynamically. That requires often functions as first class objects. Zope templates include (but not limited in future) DTML and ZPT showing a very high level of flexibility, which on the other side leads to building frameworks on a top of Zope frameworks, like CMF and Silvia.
That's why Zope is implemented on a multi-paradigm generic programming language Python, which paradigm list includes: OOP, functiona programming, dynamic typing and scripting. Inside a framework you can use on paradigm or another, but you are not forced to it.
I doubt the functionality and quality of Zope could be possible to implement on PHP, which is not a generic programming language (it's a template language), which is not designed as a programming language (it's evolved rather than mathematically thought) and it's lacking an integration of paradigms (OOP and FP specifically).
I think that Python was the lowest possible level of programming language to choose for Zope. PHP suffers as a language in general. TCL and Ruby lack in paradigms. Perl is too obfuscated. Java, C++ are compilable (not scriptable) and thus are not flexible (apart from lacking the FP paradigm).
Alternatively it could be one of "more pure" functional (and still with OOP) programming languages, like Lisp, OCaml or Haskell (or even Mozart). But those languages typically lack practical libraries and there are fewer programmers to code on them. So, the choice of Python was the best one, from both theoretical and practical stand point.
Slashdot Mirror
As for the goverment choice for KDE against GNOME - I have no problem with it (again if I have my own choice). I don't like personally KDE, but I have no problem if the goverment will use it.
99% of typical Microsoft users do not know any difference between Linux and Open Source. And they don't know what are Open Standards. You can spend one more centure educating them that linux is just a kernel, there are many distros on a top of that kerne and there are other kernels as well. Or you can spend some more reasonable time advocating Linux as a concept keeping other Open Source" in mind. What do you prefer?
I know, some BSD funs may dislike it. But in many cases it's really easier to approve "Linux in general" and later specify that sometime it can be BSD :)
Well, historically Linux was stealing a lot open source applications from BSD world. And that time for many managers Linux was one of BSD kinds. Now time is changed.
Besides, 100% of Linux is not a monoculture. There are dozens of distros. There is a competition between Gnome vs KDE vs misc managers. There are various programming languages to code the same projects. Well, even inside Perl there are always more than one way of doing the ame thing (according to Laryy Wall, the creator of Perl). Finally, there will be always PostgreSQL vs MySQL and Emacs vs vi (and even GNU/Emacs vs Xemacs for Emacs winners) - what kind of monoculture do you see here?
By the way, all those "one vs another" things are OS agnostic. If the decision maker will open the mind enough to get into Linux - in no time the person will notice that all those "vs" are the same inside BSD.
But if you will bring "Linux vs BSD" right to the lobbying process, those non-techs will decide: "OK, let's them at first decide Linux or BSD and THEN we'll see if it's worthy against Windows!" - you don't want THAT decision, don't you?
My observation shows that the most stable ones usually are those where we rarely install anything besides a single application (or suit). If either Office or Photoshop or Visual Studio or MS SQL Server is the only one which is installed - than chances of crashing are bellow the average. Two different applications and it's a big chance that we've got a problem. That makes a correction about Windows being an OS for a personal computer - it for one personal application. I think that Microsoft's QA is testing it's operating system mostly installing one of MS applications. Bad QA.
Another source of problems we are usualy having on computers with 3rd party hardware drivers. You can blame 3rd party vendors, but why I don't have such problems with the same hardware on Linux? Oh, I know why: b/c with Linux the vendor has free (often) support of developers and QA from around the world. In case of Windows the vendor is on its own doing QA often without any support from Microsoft.
We can also talk more about:
- NTFS - old FS, too bad nowaday comparing to Ext3, XFS, JFS and ReiserFS. What's wrong to get a modern FS if its more stable?
- why can't I make a page file in the raw partiotion?
- registry database is defenitely not designed for situations when you install/uninstall/reinstall many applications often. Your system will die or become slow soon if your job is installation/uninstallation/reinstallation of new software.
- Windows updates are supposed to improve my system, not to crash it. But even after failing, deinstallation of the recently applied update can crash the system as well. And nothing you can do about it.
- Windows is still using Netbios in LAN. Bad protocol, bad implementation. Wins server is too bad, especially in cache updates. Lmhosts must die already by now. Why not using DNS inside the LAN?
- Why the whole system must be locked if Explorer is trying to connect to the slow LAN server or a remote web folder?
- Eventlog very rarely gives any useful information. There should be a better control over debugging levels. Otherwise you feel yourself blind when it's crashing.
Don't tell me that I should not tinker my OS. tell it to ma&pa - home users. In a corporate environment you've got a documentation from 3rd party vendors requiring tinkering. And same level of tinkering doesn't crash the OS if ti's designed for that - I mean Linux.Meme? I think that it's meme only when you do everything only if Microsoft said you to do. But the real life is not what Microsoft was planning. That's why people more and more often prefer the OS that is designed for being owned by net/sys-admins, not just licensed. By being owned I mean being controlled as required.
Also, you can check the configuration of netfilter in your Linux kernel, or even the netfilter source code: less than 20% will be about NAT.
So, where did you get your knowledge about "80% of a firewall is NAT?"
2. If I would go with pure XSL solution that it will be Cocoon - many developers know it so it would be no problem to make a team. Cocoon has a very stable support from one of open source leaders: Apache. And Cocoon has the best XSL-oriented design for being used in content-processin applications. However, I don't think that "pure XSL" solution is the way to go. XSL doesn't cover everything. It's good for dynamic content processing, while typically I need a content management. How kristalis solves that problem?
3. The most important part missed in XSL is that functionality (as it's applied to a content) should be managed as a part of the content in a same dynamic way. That's why in Zope Python is a very essential part of DTML and ZPT. I know that some functionality can be implemented in XSL, but people who have some experience of building a very complicated systems with XSL can confirm: it's a pain in the neck. How is this issue addressed in Kristalis?
4. XSL is not designed as a general programming languages. But all attempts to extend it with non-scripting languages will almost always fail. You need a flexibility of a scripting language to embed it to tags on the fly. At the same time Perl and Tcl are not OOP enough, nor they are expressive/readable. That's why the choice is coming to Python in Zope. Can you compare it in details to Kristalis?
If you have to fix some existing PHP application - go for it. But if you have a chance to start the webapp over then do your best to make sure that your choice of Zope will be approved. Well, it's easier every day as Zope is gaining more users and success stories.
This is the excelent example of a comment from people who use OSX not because OSX is good (which is not bad), but because such people hate Windows.
I use all three desktops, Windows, (various flavors, including XP), Mac OS (including 9 and X) and Xfree (including GNOME, KDE as well as simple windos managers).
Standard disclaimer: IANADP (I am not a desktop programmer) - I don't do it for living, so my mind is not corrupted by desktop programming. Although, I still do it rarely and also although, I do software development for database, network and system management, for data and knowledge processing and for web-based content management, so I sue desktop as a end-user a lot (I mean really a lot). For example, I cut-n-paste not only text, but alose images and vector diagrams.
What I have noticed: all desktops have their own good and bad features. OSX reminds me Windows 3.1 as it's same unstable, but in a very modern themed. XP is another attempt of stylishing, although the real value of XP is in a combination of much longer debugged history (Windows today is much more stable) and still old bad misfeature of linking desktop parts to the kernel (Explorer, IE and CIFS are parts of desktop environment, right?). XFree desktops are the most stable unless you use OpenGL - a simple xscreenserver in its 3D theme can lock all your PC so much CPU it uses when something wrong. For end-users all three are the same.
Well, all said above about stability of desktops is not said about the underlying OS. Linux and BSD are the most stable OSes. OSX is the second (when it's not about GUI), Windows is the worst (we all know why).
Also, the stability of your work is a function of your applications. Specifically, how they are good with OS and do they still deliver features enough for your job on that OS. OpenOffice under Linux is the most stable Office application with still good enough feature set to my job.
Conclusion: if you really care about your job to be done than most likely you will never criticize some OS in a zealotish manner - instead you will always compare good and bad sides of both (or three) OSes.
What's the point: the underlying PHP language is one of problems of PHP, but not the ony one. I can list them here with the solution I propose (Zope):
- Language: sure Python is better choice for a readable code and OOP. In my personal opinion, Python is the best among imperative languages. I love using FP style in Python, getting the smaller code doing more and still very expressively.
- Templates: Even DTML in Zope beats PHP. As for ZPT - after learning it you cannot go back to PHP. Never.
- Namespaces: ZPT has a very excelent design and implementation of namespaces. All accordingly to W3C standards and traditions.
- Content management: in PHP you still have to write it (or find it, or buy it). With Zope you have already got it, working, convinient, secure and well designed. CMF with Plone improve it even more bringing in workflow management and collaboration.
- Database persistence: even if you don't care you already have it ZOPE. ZODB is object, transactional, reliable DBMS. If you care to integrate it with you existing SQL applications than you can use Adaptive Storage (APE) which gives you a transparent way to do it, similar to VFS+fstab give you filesystems in Linux.
I encourage disappointed PHP people to learn Zope. Even if your clueless boss will dictate you the choice of Sun MicroSCOftsystems, the value of concepts in Zope you learn cannot be overestimated - you can reuse patterns from Zope in all other other technologies, like ASP.IN FP languages, like Lisp, the code is the data: there is very well defined reflection, I can construct new functions and manipulate existing ones as they are first class objects. Same/similar situation is in ML and Haskell.
Well, among traditionally imperative programming languages there are more and more cases of "the code is the data" paradigm as well. First of all in iterpreting (scripting) languages, like Python, Perl and Tcl. You can construct the text of the function and "eval()" it. The problem is that in FP languages there is a very well designed math model for it protecting you from many errors (you construct real functions, not a text for for functions). In scripting languages it's more like a hack leading to many errors in a similar way as C pointers and Fortran GOTO operators.
As for the part of you comment where you tells that BDB is more user friendly than PostgreSQL, I think you are trolling here. Unless you want to explain yourself.
Why BDB? There is PostgreSQL if you love BSD license. And there is MySQL if you prefer GPL. And there is ZODB if you need OO layer on a top of BDB.
However, it is possible that Sun could write a whole new database application using BDB as the backend; or, Sun could write a layer for storing word processing, spreadsheet and other kinds of documents in BDB, affording some cool features that we don't get with flat binary files (which suck).
What kind of idiot would decide to do so? Only one who doesn't know (or doesn't want to know) about even one database I have mentioned above.
As it is now EU is not capable to begin any serious wars. Not from military capability prospective - such decision would be politically impossible in EU. China is also not that stupid to through nuclear warheads here and there - they realize that that would be the end for all of us.
The problem is that US administration is driven by corporations, currently by those who is benefiting from any military race. And there is no way to stop them.
Elvis Presley said: 'My voice alone is just an ordinary voice. What people come to see is how I use it. If I stand still while I'm singing, I'm dead!'
Close to later re-phrased:
"It's not my voice. It's what I am doing with it."
Back to the topic:
IMHO if my HDD is used just as a big CD - it's dead.
Apple involved? Wait for iKB, iMB and iGB - just like iTunes and iMac.
I guess Apple may change it a bit: iMB and iGB - just aka iTunes and iMac.
I asked my friends in other companies and people in mail-lists - the situation is always the same. And it's sad to compare IBM RS6K and HPPA equipment - people working with it are surpised hearing our statistics and complains.
It doesn't have to be THAT expensive if it's not more reliable than Intel same-performance servers. Or it doesn't have to be THAT unreliable for THAT price ans speed. Hmm, i am not sure if I need faster, but still expensive and THAT unreliable servers.
I guess someone should find the comparison between metro areas (no countries) to show you that NYC and LA are still worse than a dozen of other comparably same size cities. Of course per capita - that what counts by regular people comparing politicians.
Besides the car I have to get a driver license and I have to renew it from time to time. If I get too many penalty points or if I am noticed in one-time serious traffic violation my driving license can be suspended.
Same thing should be for comupters (AND networks) at the moment of connecting them to ISP:
- each computer (or a whole network) must have a license to be connected to Internet;
- periodically (as often as it's appropriate) safity and environmental checks must be take care:
- the computer must be protected in terms of ports opened and mail filters installed;
- the nightly based cron procedure must do the check and alert ISP if anything wrong found;
- from time to time (weekly or so) ISP must scan clients from outside;
Besides my computer, my Internet and PC end-user skills must be licensed:if I don't know how to update my OS or to install a security patch on it then I cannot be licensed;
if my PC is noticed as a source of virus/attacks and it's proven it's been cracked/infected than I've got my penalty points and have to pay a small fine (bellow a hundred of $);
if I am noticed as distributing viruses knowingly or hacking myself then my license should be suspended - I have to get the end-user class again, renew my license and pay a big fine (thousands of $); IMHO it will improve overall Internet safity (imagine how much less there will be opened port and unpatched computers) and accelerate the whole national economy (imaging how many "mechanics garage" companies will rise their revenue!). It will open many new IT jobs and improve exisiting ones (now my boss cannot tell me "I don't belief some one may crack us - we don't have any useful information"). By the end of the day companies will actually safe money as they spend too much now to fight security in such insecure world.
Also it will improve the competition on the market as people will prefer more secure OS to be installed on their PCs. Oops, Bill gates may hate it and lobyy against it. But I still love the idea.
For 70GB HDD, which is essentially equal to 100 CDs, and the statistical chance that the space has been reused 100 times per HDD lifetime (quite reasonable for let's say 3 years) you will be charged 10,000 times of what you would be charged per CD.
Not bad.
Interesting. Any links or your own words explaining it in deeper details? Thank you in advance.
Lisp and Python has both byte-compilation and interactive-loop interpretation. If you haven't tried yet the interpretation mode in Lisp that doesn't mean it doesn't exist in Lisp. It's even described in most of Lisp books I've seen. It's very convinient when you need a sort of interractive debug mode.
I agree that Python byte-compiled code is still not as fast as C. However, in most of application (if they are not mission critical kernel drivers or DB engines) you don't really need to squeeze that performance. But I would not say that Python bytecode is always THAT slower than Lisp bytecode.
I think you exagurate a lot here, when you use the word many to advocate C, and the word some applying to Python and other interpreters. I think that in most of projects (by amount of projects and by amount of programmers assigned to) good interpreters (like Python or Lisp or Erlang) are more preferable as they let coding in safity, they let coding fast and they cut the cost of further maintainance.
Tasks that rquiring small size or fast execution are very rarely cases, such as kernel drivers and DB engines. In most applications C is overkill in terms of compactness and speed. In some programs C is the best choice.
Modern interpretors are also more effective than before. For example, check Zope content management and application server and its companion ZODB object database. Both are written in Python. Zope when running is comparable by size to full-loaded Apache (vs typical Java based application servers), while by speed it's not (much) slower than JBoss.
Same good things can be said also about interpreters of Lisp and Erlang, another full-featured generic programming languages used in many commercial applications.
Very interesting logic: every open source project must be implemented in PHP.
Zope is a seriously designed application server. The major concept of Zope is to separate aspects. OOP is important concept but it is not the concept theat the programmer is forced to use exclusively. Also, in Zope the stateful and stateless content is managed dynamically. That requires often functions as first class objects. Zope templates include (but not limited in future) DTML and ZPT showing a very high level of flexibility, which on the other side leads to building frameworks on a top of Zope frameworks, like CMF and Silvia.
That's why Zope is implemented on a multi-paradigm generic programming language Python, which paradigm list includes: OOP, functiona programming, dynamic typing and scripting. Inside a framework you can use on paradigm or another, but you are not forced to it.
I doubt the functionality and quality of Zope could be possible to implement on PHP, which is not a generic programming language (it's a template language), which is not designed as a programming language (it's evolved rather than mathematically thought) and it's lacking an integration of paradigms (OOP and FP specifically).
I think that Python was the lowest possible level of programming language to choose for Zope. PHP suffers as a language in general. TCL and Ruby lack in paradigms. Perl is too obfuscated. Java, C++ are compilable (not scriptable) and thus are not flexible (apart from lacking the FP paradigm).
Alternatively it could be one of "more pure" functional (and still with OOP) programming languages, like Lisp, OCaml or Haskell (or even Mozart). But those languages typically lack practical libraries and there are fewer programmers to code on them. So, the choice of Python was the best one, from both theoretical and practical stand point.