The blog entry describes how people are constantly forced to work longer and longer hours, to be able to meet a deadline. The project starts off as being on track, but then they are forced to put in 6 day weeks. This will make people tired and inefficient, and the project will still only be barely on track, and they then get forced to work 12 hours a day, six days a week.
At that pressure, emplyees that can leaves for better jobs. A sure way of getting rid of the good people. The rest will be so overworked and tired, that everything they do will be sloppy and bad. People will get grumpy and nobody will enjoy work, thereby getting even less done. And the project will fall further and further behind schedule.
People shouldn't *sue* EA, they should *leave* for greener pastures. If they can't, they should complain to their boss' boss that they are forced to be inefficient by crappy management. Explain to them that nobody gets richer by having a company full of tired people too sleepy to get real work done.
And I have a suggestion for a christmas present for the EA managers: One copy each of "The Dilbert Principle". It uses simple language, and has many easy to understand illustrations. Of course, you may have to write "This is you" and draw arrows pointing to the boss for clarity, but it may very well be worth it.
Although of course wildly outdated even when it was published (as all useful computer books always are) a good book explaining the basics is never wrong. And the basics still are the same. There is still is loads of information in these old books that would be useful to anyone getting into computers, surprisingly enough...:-)
I held a course in TCP/IP in the early nineties. The part that most clearly divided the class was the net mask. People that had studied computer science, or were self-taught nerds, of course already knew binary arithmetic. They found using net masks trivial. The people who had ended up as network administrators by mistake (most of them, really) had huge problems. After holding this course a couple of times, I simply extended it with teaching everybody binary arithmetic first. That made it easier for most people.
You don't need to know how a computer works to use it anymore, but a good network manager should still know it, and a programmer won't last two weeks without understanding what actually goes on.
This is not just about science. It's even more visible in politics, which of course the primary example here was, since it was about abortion. Also it doesn't have anything to do with journalists being balanced, rather the opposite.
The idea that journalists should be fair and balanced is used as a reason for being incorrect. Nobody is ever objective, and a good journalist is not balanced, but honest. Instead of hiding opinions behind a veil of alleged objectivism, any writer should be clear about where he/she is standing in the controversy.
The idea of balance and objectivism is made worse by the idea that you should have separate people for doing news: Journalists. The result is that most of what is said in the news is said by people whos knowledge and education is in words, not in the subject that is covered. A journalist usually do not have the knowledge to say what is wrong and what is right, and is likely to spread false information even if he tries to be objective and balanced.
We need to stop listening to journalists, and start listening to people who know what they are talking about.
First of all, the US system of voter registration is completely outdated. In many modern countries you get your paper that tells you that you are eligible to vote send home to you with ordinary mail well in advance of the vote. Come on, the US goverment knows who are citizens, they know who get to vote. Why do you have to tell them first? This would get rid of any need for pruning or verifying the registration lists. This would also get rid of people registering Betty Boop for crack...
Secondly, you need a better paper trail. As it is now, the federal goverment can't tell the states how the voting should be done, with the effect that you have a plethora of different types of voting machines, many of them either confusing or insecure. But why should not the federal government be able to impose rerquirements on how federal voting is done? A good voting machine is simple: A computer with a touch-screen where you select who you want to vote for, get the answer "Are you sure?". When you are sure it prints out a paper. You confirm that the paper says the same thing as what you actually pressed on the screen, and you put it in a ballot box.There! Now you have an electronic voting machine that is extremely simple to use, and, most importantly, provides a paper trail, so that if the machines are called into question, you can recount the ballots manually. Machines like this helped verify the referendum in Venezuela as valid. Machines like this would have made it very simple to validate the votes both 2000 and 2001.
And, yes, as so many others, I would make the system proportional, and get rid of the ancient system of voting for electors. It was a good idea in the end of the 18th century. It's not a good idea now.
And that's just the election procedures. Just wait to I change the rest of US politics!
Oh, right, I don't get to change anything, I just get to rant. Ah well, better than nothing!:-)
Is open source-politics a good idea? Then we first has to ask us "what does *that* mean"?
If we start with the ideas mentioned in the above articles, then it has a couple of examples of politicians using the internet. Internet allows politicians to increase their efficiency. They can now answer peoples questions without meeting them in person, when they have time to answer them (not when the person has time to ask them) and let many others know the answer. All of this was possible before, but not at the same time. Now it is!
It also allows politicians to efficiently get to know the opinions of the supporters. For example, Swedish politician Lars Leijonborg (http://lars.leijonborg.se/) have for a couple of years now regularily made on-line polls that subscribers to his newsletter can answer (he calls the "advisers"), and comment on. He can that way not only see which action his supporters prefer, but he can see their comments. You get a direct line into the leader of a Swedish political party, to give your opinions of current issues.
Is this good? Of course! It's great! Just like internet and e-mail otherwise have flattened hierarchies, it now does so with politis. This was a revolutionary concept in company management ten years ago. Now it has reached politics. Kick out the middle men and talk directly to people in power. That's great! We need more of that!
But is it open-source? Nah, of course not. That's just some people taking a trendy buzzword and applying it onto something they want to promote. It should probably be called email politics, or internet politics, since there is no more openness than in any other politics, and there is no source in sight.
This is exactly the long slope of the lurning curve that was a problem with Zope2. Zope3 fixes it, and replaces it with a slightly higher, but well documented, threshold.
So nom you will not have to do it again with Zope3.
1. Well, as mentioned above, there is no need for an object relational layer when you have an object store. There is also good relational integration when you need to talk to external datasources. But unless you need to talk to external datasources, storage is automatic. You don't have to think about it. (Although if you want to you can).
2. Zope has transactional control. Most of it is simply automatic, you don't have to care. If something fails, the transaction backs out. You can start new substransactions, commit, and back out as you wish, should you need it. You can also easily hook into the transaction control so that you can make external resources transaction aware. But as mentioned, usually, this is already taken care of.
3. I don't even know what that means. Can I call code in other containers? Sure. Is there a naming service, ehm, no...? I think this is simply yet another thing that gets meaningless in Zope. You don't need it.
4. CORBA support? Why on earth would you need that? SOAP support, yes, you need it if you need to talk to another webserver who requires you to use SOAP to talk to it. There is no problem implementing this, and many have, although there is nothing automagic included in Zope. All in all, it is daily shown that in fact, Zope does NOT need SOAP support to compete with J2EE, because it *is* competing with J2EE, and sucessfully so, and most people using Zope never uses SOAP at all. But if you need it, you can have it.
"What organizational requirements are met by Zope?"
My response to that is: Huh? I don't understand the question. It's a development platform! What organizational requirements are met by C++, or for that matter by MS Office??
"What kinds of sites does it make easier?"
It doesn't make "sites" easier. It's a web application development platform. It makes almost all web applictaion developemnt easier. If you are looking for software to develop sites, you probably want a CMS of some sort. There are several ones that runs on Zope, the most popular one being Plone, and CPS being a more advanced CMS (neutrality warning: I'm a CPS developer;) ).
"In what ways easier? Development? Maintenance?"
Mainly development. Deployment and scaling is usually quite easy too.
"How much easier?"
On a scale from 1 to 10: 9. Zope3 I would expect to score 11 to 12.;) It is *very* quick to develop applictaions with, once you understand the concepts.
In some ways you are right. Zope2s learning curve has an extremely low initial treshold, making it very easy to start. But after that threshold there is a slope that for a long time seems to never end (it does, eventually). Compare this with for example Python, which you feel that you have mastered after a day, and it only bites you in the bum once of twice after that.
Zope3 on the other hand has a much higher initial threshold than Zope2. You need to spend a couple of days going through the tutorial and the first chapters of the Zope3 books. And there are a couple of concepts that need to fall in place.
But after that, the slope with Zope3 is much lighter. Also, the delevoplemnt documentaion for the framework is much better and this should help you be able to use new parts of that framework quickly(just as finished CMSs) more quickly.
Python, Scriptlanguages and non-script languages.
on
Zope X3 3.0.0 Released
·
· Score: 1
Python is called a "script language" because it provides an interpreter. Java and C does not. This of course sucks, an interpreter is VERY handy for testing stuff.
If you don't want to get ridiculed, call pythons interpreter a "just-in-time bytecode compiler". because that's what it is. Instead of running a separate program to compile code into byte-code as you do with Java, Python does it on the fly.
There are also lots of cool extras, like Psyco, which is another jit-compiler, but it will actually compile to machinecode. Not ony that, it will precompile common results of functions, so instead of calling the method it will just look up the result if you already called the method with these parameters. This can actually sped up many algoritmic functions much more than it would rewriting them in C. (Psyco is usable with Zope, btw, and will speed up for example Plone a lot).
Other cool extensions are Pyrex, which is somewhat of a mix of C an Python. It let's you use Python-syntax to make C-code.
The whole point of all this is that the difference between compiled languages, byte-compiled languages and interpreted languages has gone away. The borders between these are now very fuzzy and the speed differences decreasing to the point that interpreted languages in some extreme cases are FASTER than compiled languages.;)
So if somebody balks at script langauges, they are not up to date on language development. If they are smart people: Get them up to date. If they are not: Just don't mention the s-word (that is, scripts).
Well, first of all, DTML was way ahead of it's time when it came. It kicks the whiney little butt of inline-escape-to-programming-languages environments like PHP and ASP.
Now, admittedly, DTML had it's share of problems. These were fixed a long time ago with the introduction of ZPT. I haven't touched DTML in 18 months (and I was a slow converter).
ZopeX3, as this is about doesn't even support DTML. So there.;-)
ZODB is admittedly hard to scale if you have heavy WRITE loads. A classic error is storing page-view counters in the ZODB. Not good for performance.;) However, for heavy READ loads, it is eminently scalable. This is of course 99.9% of all websites...;-)
No, this is incorrect. I don't know anything about Drupal, but PHP is a language specifically made to create dynamic websites. It is not comparable to Zope in any way. PHP is comparable to Python only. Zope is written with Python.
You are right, Zope is not an alternative to PHP/MySQL. Zope is a complete platform for developing web applications. PHP/MySQL is a language for making dynamic HTML and an relational database server. It's a bit like saying that a complete electrified woodshop is not an alternative to an axe and a hacksaw.:-)
Well, it isn't, but when you wanna build yourself a nice set of furniture I bet you'll be much happier with the woodshop than with the axe and hacksaw.
I definitely do not agree with this. Any project except small ones end up using a framework, unless the developers are crap or very determined to stay on the project forever. You either make your own (usually half-assed) framework, or you use one that exists already.
I agree that frameworks are a dime a dozen, but that is because most projects who made their own framework today usually ends up open sourcing it, just as most small webcompanies who developed a CMS to use for their sites ended up trying to sell their hack as an "enterprise solution".:-)
Zope is different, becuase many people are using it, and it is not crappy or half-assed. This admittedly is partly because it was opens sourced early, but it is mostly becuase it uses several good radical concepts that vastly simplifies development. It is a genuinly object-oriented environment using an object-store instead of the SQL-server common for most other frameworks. And if you don't understand why that is so fantastic, then it's because you haven't tries it.;)
Zope3 is a rewrite to address some issues of Zope2s framework utilizing even more new concepts. For example, the Zope 3 component architecture uses "adapters" to merge in the concepts of aspect-oriented programming into the oo-framework. It looks very nice, so far, and I look very much forward to get to do a larger project with it.
Well, actually, that pretty much describes what Zope3 is. It's not J2EE like as in "big, bloated, and no fun to use" but it is J2EE-like in that is is a framework to "develop component-based multitier enterprise applications" which is the description of J2EE.
...this problem has been well known for a long time. And it's not only that they compress the hell out of everything, the last few years they are even pushing the volume so hard that it's actually starting to clip. Not enough to be obvious, especialy not on more heavy music where the guitars are already heavy with distortion, but still.
No, of course louder is not always better. Often the record company in an effort to make the album loud force the mastering to use so much compression that the life is sucked out of the songs, and what surely once was a nice intro with a powerful snare drum that cracks becomes a puny intro with the snare drums no longer is loud, but just goes 'piff' and forces down the volume of the rest of the instruments, thanks to the compression.
It's all quite silly, IMO, especially since the radio that they want to be loud on will add their own heavy compression, so the heavy compression they put on the album is useless anyway.
"Today, data persistence for object-oriented systems is an incredibly cumbersome task to deal with when building many kinds of applications. The developer must map objects to database tables, XML files or use some other non-OO way to represent data, destroying encapsulation completely. One solution to this problem is object prevalence."
No. Not if you store things in some kind of object store. Like the one mentioned in this article, or an OODB. This guys is not the first to do that, even though he sound like they think he is, but then, being 19, that is quite excusable.:)
...but this is hardly anything revolutionary. Sure, OGG is a nice format, and it is supposedly patent free. Until somebody actually discovers that they have a patent for something that it is using, that is.:)
The big issue is always a question of "does it work today"? And today, there are free software to decode and encode both mp3 and ogg files, and that means that they are equivalent from that standpoint. A small number of people will preferr ogg out of purely moral reasons, but for the vast majority this only means that there is yet another file format out of loads of fileformats that is supported by Apple. Which is nice, but on the whole rather uninteresting.
Apple did hardly implement this to make a point. They probably implemented it because it was easy.:) And 99.99% of Mac and Linux user won't ever notice.:)
I don't know enough about the human hearing system to say if their description of it is corrcet or not, so let's assume that it is. In that case the claim that the ear cannot calibrate itself with mp3 compressed sound is incorrect, as these signals for self calibration is claimed to be generated by resonances within the cochlea. These signals will therefore not be missing at all. What the compression algorithms are removing are sounds in the music that would be masked by these resonances anyway.
At the end of the paper the author claim to repeatedly have tinnitus (which then is not tinnitus at all, because tinnitus is *permanent*, thank you very much) despite listening to music only at low volumes. He seems to blame this on mp3 compression, despite the fact that he according to his own words never listen to that either.
The article is listed under the part of the authors web that is dedicated to "Logology" which according to the author is "the first Cyberage religion".
So, the only thing left to determine, is if this is irony, or if the author is insane. I sure don't know.:-)
Slashdot Mirror
The blog entry describes how people are constantly forced to work longer and longer hours, to be able to meet a deadline. The project starts off as being on track, but then they are forced to put in 6 day weeks. This will make people tired and inefficient, and the project will still only be barely on track, and they then get forced to work 12 hours a day, six days a week.
At that pressure, emplyees that can leaves for better jobs. A sure way of getting rid of the good people. The rest will be so overworked and tired, that everything they do will be sloppy and bad. People will get grumpy and nobody will enjoy work, thereby getting even less done. And the project will fall further and further behind schedule.
People shouldn't *sue* EA, they should *leave* for greener pastures. If they can't, they should complain to their boss' boss that they are forced to be inefficient by crappy management. Explain to them that nobody gets richer by having a company full of tired people too sleepy to get real work done.
And I have a suggestion for a christmas present for the EA managers: One copy each of "The Dilbert Principle". It uses simple language, and has many easy to understand illustrations. Of course, you may have to write "This is you" and draw arrows pointing to the boss for clarity, but it may very well be worth it.
The transport revolution that never was ran over the the dream of a household robot. Many science-fiction dreams was hurt in the crash.
"I met a girl who looks a lot like you
She does then things you do
But she is an IBM."
- Electric Light Orchestra
Although of course wildly outdated even when it was published (as all useful computer books always are) a good book explaining the basics is never wrong. And the basics still are the same. There is still is loads of information in these old books that would be useful to anyone getting into computers, surprisingly enough... :-)
;)
I held a course in TCP/IP in the early nineties. The part that most clearly divided the class was the net mask. People that had studied computer science, or were self-taught nerds, of course already knew binary arithmetic. They found using net masks trivial. The people who had ended up as network administrators by mistake (most of them, really) had huge problems. After holding this course a couple of times, I simply extended it with teaching everybody binary arithmetic first. That made it easier for most people.
You don't need to know how a computer works to use it anymore, but a good network manager should still know it, and a programmer won't last two weeks without understanding what actually goes on.
Well, maybe if he is using Python.
This is not just about science. It's even more visible in politics, which of course the primary example here was, since it was about abortion. Also it doesn't have anything to do with journalists being balanced, rather the opposite.
The idea that journalists should be fair and balanced is used as a reason for being incorrect. Nobody is ever objective, and a good journalist is not balanced, but honest. Instead of hiding opinions behind a veil of alleged objectivism, any writer should be clear about where he/she is standing in the controversy.
The idea of balance and objectivism is made worse by the idea that you should have separate people for doing news: Journalists. The result is that most of what is said in the news is said by people whos knowledge and education is in words, not in the subject that is covered. A journalist usually do not have the knowledge to say what is wrong and what is right, and is likely to spread false information even if he tries to be objective and balanced.
We need to stop listening to journalists, and start listening to people who know what they are talking about.
winamp was finally getting good. A few small glitches to iron out and that would have been it...
ah, well.
...after he has been outed. We also know that nature likes frauds that get away with it. Just like people, then.
Proving the obvious sounds like a good job. Or a good fraud...
...you should upgrade your gear as soon as you can find any weak justification for it! Righto!
We all have G.A.S. Just give in to it!
Mum, mum, look, a troll. Can I keep it, please, mum?
- First of all, the US system of voter registration is completely outdated. In many modern countries you get your paper that tells you that you are eligible to vote send home to you with ordinary mail well in advance of the vote. Come on, the US goverment knows who are citizens, they know who get to vote. Why do you have to tell them first? This would get rid of any need for pruning or verifying the registration lists. This would also get rid of people registering Betty Boop for crack...
- Secondly, you need a better paper trail. As it is now, the federal goverment can't tell the states how the voting should be done, with the effect that you have a plethora of different types of voting machines, many of them either confusing or insecure. But why should not the federal government be able to impose rerquirements on how federal voting is done?
- And, yes, as so many others, I would make the system proportional, and get rid of the ancient system of voting for electors. It was a good idea in the end of the 18th century. It's not a good idea now.
And that's just the election procedures. Just wait to I change the rest of US politics! Oh, right, I don't get to change anything, I just get to rant. Ah well, better than nothing!A good voting machine is simple: A computer with a touch-screen where you select who you want to vote for, get the answer "Are you sure?". When you are sure it prints out a paper. You confirm that the paper says the same thing as what you actually pressed on the screen, and you put it in a ballot box.There! Now you have an electronic voting machine that is extremely simple to use, and, most importantly, provides a paper trail, so that if the machines are called into question, you can recount the ballots manually.
Machines like this helped verify the referendum in Venezuela as valid. Machines like this would have made it very simple to validate the votes both 2000 and 2001.
Is open source-politics a good idea? Then we first has to ask us "what does *that* mean"?
If we start with the ideas mentioned in the above articles, then it has a couple of examples of politicians using the internet. Internet allows politicians to increase their efficiency. They can now answer peoples questions without meeting them in person, when they have time to answer them (not when the person has time to ask them) and let many others know the answer. All of this was possible before, but not at the same time. Now it is!
It also allows politicians to efficiently get to know the opinions of the supporters. For example, Swedish politician Lars Leijonborg (http://lars.leijonborg.se/) have for a couple of years now regularily made on-line polls that subscribers to his newsletter can answer (he calls the "advisers"), and comment on. He can that way not only see which action his supporters prefer, but he can see their comments. You get a direct line into the leader of a Swedish political party, to give your opinions of current issues.
Is this good? Of course! It's great! Just like internet and e-mail otherwise have flattened hierarchies, it now does so with politis. This was a revolutionary concept in company management ten years ago. Now it has reached politics. Kick out the middle men and talk directly to people in power. That's great! We need more of that!
But is it open-source? Nah, of course not. That's just some people taking a trendy buzzword and applying it onto something they want to promote. It should probably be called email politics, or internet politics, since there is no more openness than in any other politics, and there is no source in sight.
This is exactly the long slope of the lurning curve that was a problem with Zope2. Zope3 fixes it, and replaces it with a slightly higher, but well documented, threshold.
So nom you will not have to do it again with Zope3.
1. Well, as mentioned above, there is no need for an object relational layer when you have an object store. There is also good relational integration when you need to talk to external datasources. But unless you need to talk to external datasources, storage is automatic. You don't have to think about it. (Although if you want to you can).
2. Zope has transactional control. Most of it is simply automatic, you don't have to care. If something fails, the transaction backs out. You can start new substransactions, commit, and back out as you wish, should you need it. You can also easily hook into the transaction control so that you can make external resources transaction aware. But as mentioned, usually, this is already taken care of.
3. I don't even know what that means. Can I call code in other containers? Sure. Is there a naming service, ehm, no...? I think this is simply yet another thing that gets meaningless in Zope. You don't need it.
4. CORBA support? Why on earth would you need that? SOAP support, yes, you need it if you need to talk to another webserver who requires you to use SOAP to talk to it. There is no problem implementing this, and many have, although there is nothing automagic included in Zope. All in all, it is daily shown that in fact, Zope does NOT need SOAP support to compete with J2EE, because it *is* competing with J2EE, and sucessfully so, and most people using Zope never uses SOAP at all. But if you need it, you can have it.
"What organizational requirements are met by Zope?"
;) ).
;) It is *very* quick to develop applictaions with, once you understand the concepts.
My response to that is: Huh? I don't understand the question. It's a development platform! What organizational requirements are met by C++, or for that matter by MS Office??
"What kinds of sites does it make easier?"
It doesn't make "sites" easier. It's a web application development platform. It makes almost all web applictaion developemnt easier.
If you are looking for software to develop sites, you probably want a CMS of some sort. There are several ones that runs on Zope, the most popular one being Plone, and CPS being a more advanced CMS (neutrality warning: I'm a CPS developer
"In what ways easier? Development? Maintenance?"
Mainly development. Deployment and scaling is usually quite easy too.
"How much easier?"
On a scale from 1 to 10: 9. Zope3 I would expect to score 11 to 12.
In some ways you are right. Zope2s learning curve has an extremely low initial treshold, making it very easy to start. But after that threshold there is a slope that for a long time seems to never end (it does, eventually). Compare this with for example Python, which you feel that you have mastered after a day, and it only bites you in the bum once of twice after that.
Zope3 on the other hand has a much higher initial threshold than Zope2. You need to spend a couple of days going through the tutorial and the first chapters of the Zope3 books. And there are a couple of concepts that need to fall in place.
But after that, the slope with Zope3 is much lighter. Also, the delevoplemnt documentaion for the framework is much better and this should help you be able to use new parts of that framework quickly(just as finished CMSs) more quickly.
Python is called a "script language" because it provides an interpreter. Java and C does not. This of course sucks, an interpreter is VERY handy for testing stuff.
;)
If you don't want to get ridiculed, call pythons interpreter a "just-in-time bytecode compiler". because that's what it is. Instead of running a separate program to compile code into byte-code as you do with Java, Python does it on the fly.
There are also lots of cool extras, like Psyco, which is another jit-compiler, but it will actually compile to machinecode. Not ony that, it will precompile common results of functions, so instead of calling the method it will just look up the result if you already called the method with these parameters. This can actually sped up many algoritmic functions much more than it would rewriting them in C. (Psyco is usable with Zope, btw, and will speed up for example Plone a lot).
Other cool extensions are Pyrex, which is somewhat of a mix of C an Python. It let's you use Python-syntax to make C-code.
The whole point of all this is that the difference between compiled languages, byte-compiled languages and interpreted languages has gone away. The borders between these are now very fuzzy and the speed differences decreasing to the point that interpreted languages in some extreme cases are FASTER than compiled languages.
So if somebody balks at script langauges, they are not up to date on language development. If they are smart people: Get them up to date. If they are not: Just don't mention the s-word (that is, scripts).
Well, first of all, DTML was way ahead of it's time when it came. It kicks the whiney little butt of inline-escape-to-programming-languages environments like PHP and ASP.
;-)
;) However, for heavy READ loads, it is eminently scalable. This is of course 99.9% of all websites... ;-)
Now, admittedly, DTML had it's share of problems. These were fixed a long time ago with the introduction of ZPT. I haven't touched DTML in 18 months (and I was a slow converter).
ZopeX3, as this is about doesn't even support DTML. So there.
ZODB is admittedly hard to scale if you have heavy WRITE loads. A classic error is storing page-view counters in the ZODB. Not good for performance.
No, this is incorrect. I don't know anything about Drupal, but PHP is a language specifically made to create dynamic websites. It is not comparable to Zope in any way. PHP is comparable to Python only. Zope is written with Python.
You are right, Zope is not an alternative to PHP/MySQL. Zope is a complete platform for developing web applications. PHP/MySQL is a language for making dynamic HTML and an relational database server. It's a bit like saying that a complete electrified woodshop is not an alternative to an axe and a hacksaw. :-)
Well, it isn't, but when you wanna build yourself a nice set of furniture I bet you'll be much happier with the woodshop than with the axe and hacksaw.
I definitely do not agree with this. Any project except small ones end up using a framework, unless the developers are crap or very determined to stay on the project forever. You either make your own (usually half-assed) framework, or you use one that exists already. I agree that frameworks are a dime a dozen, but that is because most projects who made their own framework today usually ends up open sourcing it, just as most small webcompanies who developed a CMS to use for their sites ended up trying to sell their hack as an "enterprise solution". :-)
Zope is different, becuase many people are using it, and it is not crappy or half-assed. This admittedly is partly because it was opens sourced early, but it is mostly becuase it uses several good radical concepts that vastly simplifies development. It is a genuinly object-oriented environment using an object-store instead of the SQL-server common for most other frameworks. And if you don't understand why that is so fantastic, then it's because you haven't tries it. ;)
Zope3 is a rewrite to address some issues of Zope2s framework utilizing even more new concepts. For example, the Zope 3 component architecture uses "adapters" to merge in the concepts of aspect-oriented programming into the oo-framework. It looks very nice, so far, and I look very much forward to get to do a larger project with it.
Well, actually, that pretty much describes what Zope3 is. It's not J2EE like as in "big, bloated, and no fun to use" but it is J2EE-like in that is is a framework to "develop component-based multitier enterprise applications" which is the description of J2EE.
Nono, it IS working correctly with Mozilla, that is: Not at all. It is IE that is broken. Just don't use it.
...this problem has been well known for a long time. And it's not only that they compress the hell out of everything, the last few years they are even pushing the volume so hard that it's actually starting to clip. Not enough to be obvious, especialy not on more heavy music where the guitars are already heavy with distortion, but still. No, of course louder is not always better. Often the record company in an effort to make the album loud force the mastering to use so much compression that the life is sucked out of the songs, and what surely once was a nice intro with a powerful snare drum that cracks becomes a puny intro with the snare drums no longer is loud, but just goes 'piff' and forces down the volume of the rest of the instruments, thanks to the compression. It's all quite silly, IMO, especially since the radio that they want to be loud on will add their own heavy compression, so the heavy compression they put on the album is useless anyway.
"Today, data persistence for object-oriented systems is an incredibly cumbersome task to deal with when building many kinds of applications. The developer must map objects to database tables, XML files or use some other non-OO way to represent data, destroying encapsulation completely. One solution to this problem is object prevalence."
:)
No. Not if you store things in some kind of object store. Like the one mentioned in this article, or an OODB. This guys is not the first to do that, even though he sound like they think he is, but then, being 19, that is quite excusable.
...but this is hardly anything revolutionary. Sure, OGG is a nice format, and it is supposedly patent free. Until somebody actually discovers that they have a patent for something that it is using, that is. :)
:) And 99.99% of Mac and Linux user won't ever notice. :)
The big issue is always a question of "does it work today"? And today, there are free software to decode and encode both mp3 and ogg files, and that means that they are equivalent from that standpoint. A small number of people will preferr ogg out of purely moral reasons, but for the vast majority this only means that there is yet another file format out of loads of fileformats that is supported by Apple. Which is nice, but on the whole rather uninteresting.
Apple did hardly implement this to make a point. They probably implemented it because it was easy.
I don't know enough about the human hearing system to say if their description of it is corrcet or not, so let's assume that it is. In that case the claim that the ear cannot calibrate itself with mp3 compressed sound is incorrect, as these signals for self calibration is claimed to be generated by resonances within the cochlea. These signals will therefore not be missing at all. What the compression algorithms are removing are sounds in the music that would be masked by these resonances anyway.
:-)
At the end of the paper the author claim to repeatedly have tinnitus (which then is not tinnitus at all, because tinnitus is *permanent*, thank you very much) despite listening to music only at low volumes. He seems to blame this on mp3 compression, despite the fact that he according to his own words never listen to that either.
The article is listed under the part of the authors web that is dedicated to "Logology" which according to the author is "the first Cyberage religion".
So, the only thing left to determine, is if this is irony, or if the author is insane. I sure don't know.