Re:Before all the flamers get in.
on
Qt On DirectFB
·
· Score: 3, Insightful
Why do ex-windows users keept claiming this? It is utter nonsense. X window has nothing to do with slow spread of unix on the desktop.
Do you really think business users are disturbed by too little FPS for 3 D shootere?!?
Yes, Motif is ugly. But it is not ugly because of X, but just because it is ugly. X is just a way to draw on a screen (locally or remotely). It is more than fast enough for any 2D needs one might have nowadays, even through a LAN. There are enough alternatives to Motif nowadays, but no single de facto standard has established itself yet. Do you think that by removing X and its network transparency, some elegant GUI toolkit suddenly establishes itself as de facto standard? Also on a direct FB it is possible to run many GUI toolkits.
There are more than enough hardware supported (2D) X windows drivers. How can you claim that X was designed for and ran well on a 8 Mhz 68000 CPU, and does not run well on current machines which are 100 times as fast?
As for people wanting desktop machines: "people" in fact want windows machines, because that is what they know and most are not interested in computers. However most incentives for introducing unix on the desktop in companies is to save cost, therefore the IT department/architecture forces the users whether they like it or not. Thus "thin clients" are introduced. Not the pure thin clients that are only a network terminal, but those that have the apps that are used most (office suite) installed locally and all other apps remotely, and running remotely via X.
That is a great solution, because it saves the IT department loads of local support and configuration. New version? No need to distribute software to 10000 desktops, just upgrade on a couple of servers. Why do you think even Microsoft has introduced "terminal server" after years of denial? Only when citrix got really popular they had to discover that many clients (i.e. large corporations) do want a mixture of desktop and thin clients.
Summary: it is an illusion and unfounded claim that X somehow prevents a migration to desktop unix. It is not because of, but rather in spite of X windows that migration to desktop unix is slow.
Network transparency is very important
on
Qt On DirectFB
·
· Score: 1
Especially in companies. At most places in companies where unix/linux is used, it is used by servers and thin clients. The network transparency is absolutely crucial. Even MSFT tried to add it to its GUI using terminal server.
Also at home it is useful to many. All the people I know that run linux or freebsd at home (except those that just try it out occasionally) have a linux or freebsd server and a windows or linux/freebsd desktop. Most apps are run through X, just 1 or 2 games that need more speed are run locally.
I think everyone how "grew up" on unix uses it this way. Only people that grew up on windows and just recently moved to linux are used to setting things up so that all interactive apps run directly on the desktop.
Which does not necessarily mean it shall be installed on 80%. In fact that number seems like a wild guess and very improbably to me.
If installing windows in vmware, be sure they have to pay the full price. In that case it would be impossible that the "linux" deal would be only a bit more expensive as the "windows" deal would have been. It should have cost double at least in that case.
However, it is well possible that they use a few servers with vmware GSX (the server variant) to which many PC's may occasionally connect via the vmware GSX client to run a legacy windows application. In that case a few windows licences would do.
Why does the top want XYZ? Because a customer (in the end a consumer) wants to buy it.
Who shall be left to buy, if all jobs except for the 3 top guys are "outsourced"?
Also: the consumer wants XYZ, he does not care if it is produced by a company in the US, Europe or India or China. So, the CEO of the US company shall loose his job at latest when the company ceases to exist, or is taken over by an Indian company.
There is nothing that makes the top guys more immune to this problem than others. It just will take a bit longer.
If Asian countries don't implement laws to effectively fight spam, this might be a way to force them. If people and businesses can no longer send us e-mail because personal servers, but also ISP's simply blacklist an entire country, I think it might have some effect.
Using relays based in other countries won't be possible either, since relays are also blacklisted.
90% of spam I received came from Asia (korea, china etc) until I added the following to my postfix access config file:
202 554 All sites from Asia-Pacific NIC blocked due to excessive SPAM 203 554 All sites from Asia-Pacific NIC blocked due to excessive SPAM 210 554 All sites from Asia-Pacific NIC blocked due to excessive SPAM 211 554 All sites from Asia-Pacific NIC blocked due to excessive SPAM 218 554 All sites from Asia-Pacific NIC blocked due to excessive SPAM 219 554 All sites from Asia-Pacific NIC blocked due to excessive SPAM 220 554 All sites from Asia-Pacific NIC blocked due to excessive SPAM 221 554 All sites from Asia-Pacific NIC blocked due to excessive SPAM 61 554 All sites from Asia-Pacific NIC blocked due to excessive SPAM
But, in the end, they are protecting the companies who fund them. And quests such as not buying CDs in order to protest the RIAA only result in more justification for the RIAA to encourage cracking down.
If people don't buy CD's long enough, the companies funding the RIAA shall go bankrupt and noone is left to fund the RIAA. Then the RIAA shall no longer have money to buy politicians and laws.
Because of this car makers have discouraged, even threatened, car dealers not to sell to foreigners, e.g. a danish dealer would risk his dealership if he sold to a german.
Now the EU has declared that illegal. The result is that the net prices (before VAT) shall be the same everywhere, so the price incl. VAT shall be much higher in denmark than in germany.
I think this is only reasonable. If some state decides it wants to have much higher taxes, the result is that the prices become much higher. It is unjust that consumers in countries with low taxes kind of subsidize those with high taxes.
So: what stops apple to offer their hardware in Hungary for the same net prices (excl. taxes and shipping) as in the US, then the consumer in hungary must pay extra for taxes and more expensive shipping. It should be up to him to decide, under those conditions, he still wants to buy.
Of course one cannot force a company to sell a product somewhere.
However I think in the case of webshops, I think a common reason is that in many countries there is still much more reliance on local dealers and importers; apple doesn't want to piss off its local dealers by competing with them by means of direct sales.
If I spread documents/e-books relying on such protection, I would not be a happy customer.
Sure they can sue makers of commercial cracking programs, but fact remains that many people can use such programs to make 'illegal' copies of my documents. Suing with the DMCA doesn't make much change in that respect.
Yes, MSFT shall not be able to keep up their rate of changing technology. They have been forced to do so and I think they shall suffer once they no longer can do this.
Reason: It has been their way of keeping competition shut out. As soon as others became able to reverse engineer their file formats, API's and network protocols, their products threatened to loose their exclusive nature and others would be able to create products that open a possible migration path. Once they (have to) stop doing this, it will become much easier for customers to move on, away from MSFT.
As long as MSFT is trying to make interoperability with office file formats difficult, it shall remain imperfect: i.e. as long as they are not forced to change their sabotage by court.
However, this only means that some (large) organisation has to be bold enough and take the first step, and say goodbye to MSFT office. And who is in a better position to do so (and force all parties that want to be involved with your organisations to adapt) than the state? It is the only organisation that can change laws and thus make this happen.
I expect that this strangling grip of MSFT Office can only be broken by governments/states to take a first step. This means that:
citizens do not have to be forced to buy expensive software to deal with the state (which is immoral IMO)
companies that want to deal with the state have to adapt and use an open file format as well
The benefits of being freed from the current hostage situation shall be enormous, and everyone shall be grateful for it, even if the initial steps may cause some inconvenience and money. Sometimes you have to be not so shortsighted plan a bit further ahead.
Ion is a descendant of PWM that is being actively maintained. I've been using it since the last year (after 15 years of twm and ctwm).
evilwm, I would say, is a close second. It happens to have exactly the bindings that I configured for twm. But I do like the idea of PWM and ION to always maximize windows and abolish the overlapping windows concept. It makes maximum use of your screen and minimizes the need to use a mouse for window management tasks.
Generally, I'm sceptic towards syntactic sugar (such as operator overloading, or even overloading in general). However, if you call generics syntactic sugar, you don't know what they are or what you are taling about.
Sure, its implementation may be in terms of what already exists (i.e. internally you still have Lists of Objects, being cast into Integers or whatever). However, the burden of typecasting is removed from the programmer into the compiler, which is very significant and not syntactic sugar at all!
C++ in its first incarnations was also just a precompiler to C, so following your reasoning one might have called C++ just syntactic sugar. Or even a compiler as mere syntactic sugar w.r.t. assembly, since in the end it is machine code that runs in both cases.
No, getting rid of typecasts, a source of many errors and mistakes, is a big step towards a more safe and better readable language.
As far as the remaining lack of stack variables: I agree it would be nice to have them. However I consider this an implementation/optimization detail. It should be up to the compiler to decide whether to use stack or heap for a particular case (though I'm not sure if current compilers can make the best decision). Kind of like the 'register' keyword in C, that hinted the compiler to store an important variable in a register. Due to progress in compiler technology it has become totally redundant and even harmful.
But don't forget that even thoug landline phone is relatively expensive, mobile is even more expensive. Especially if using it across national borders, it is obscenely expensive.
Also for data (modem, ADSL) mobile phone is not really an alternative.
I see mobile telephony mostly as fashionable, but less functional (slow data speed, very expensive).
It is a programming language (stack based, like forth), which is amazing for its primary purpose: printing.
But to call the language itself "extremely powerful" is an exaggeration. As a programming language it is quite primitive and incomprehensible, compared to more powerful languages such as C++, java, ML, yes even forth.
There is a big difference between SUN and SGI (apart from what others have said here):
The E15K is used by big companies such as banks as mainframe like or mainframe replacement computers.
The fast SGI machines were mainly used as scientific supercomputers, which is a small niche market.
The mainframe market for superservers and UNIX consolidation however is huge and no niche market at all!
Where I work (2nd bank of Switzerland) it is E15K's all the way. All software written in house (by thousands of developers) runs either on MVS or on E15K, partitioned to serparate and guarantee applications the capacity they need.
Apart from the fact that X is very good, it is absolutely crucial that Linux remains compatible with the rest of the "free world", that is UNIX and *BSD.
Remember, these are all allies against a common enemy. Going it alone in a crucial such as the GUI would be fatal for us all, and that further fragmentation be avoided.
Read/(write) access to none native filesystems such as NTFS is only relevant for dual boot systems, where it may be nice to access your NTFS partition while booted into *BSD or Linux.
But "real" systems are no dual boot systems. So you don't need it. Hardly find NTFS on floppies or CD-R or on tape. OK maybe for a hotswappable (scsi) harddisk it might have a use, but that is the only serious thing I can think of.
All other interoperability between filesystems goes via network filesystems, be it SMB, NFS, AFS, DFS or whatever. Those are the ones you need. A good and free implementation of NFS for NT might be nice, or Samba keeping up with ever changing Windows fileservers. For real use you need a directory service across platforms as well (NIS+ on windows, active directory on UNIX).
WHAT?! I could say the same thing about UNIX's user/group/world semantics, and far more defensibly. ACLs allow all sorts of useful things; I can have a log directory that's append-only except to sysadmins, for instance; or have a mail spool directory writable to only its owner and processes on the mailserver; or lots of useful things that standard UNIX semantics don't support. Who's wearing the straitjacket, again?
Fact is, like it or not (I love it) that while using UNIX you have to live with user/group/world semantics.
It may not allow the ad-hoc fine-tuning that ACL's allow, but IMO that is a good thing.
Really, ACL's may quickly become a mess. Yes you can set some special permissions and exceptions fast, without too much planning or thinking (as groups require) but after a while it is unmanageable. Why do you think that many larger sites running large Windows implementations have a policy of not allowing ACL?!?
But still, a few years ago you'd have to buy the disc again in such cases. Nowadays you just burn again from the original. 100% legitimate of course, but it is costing them turnover compared to the old days.
I can only hope that their profits shrink so fast that they cannot afford to buy laws much longer.
I hardly see any place left for "coders" these days. The word "coder" implies someone who routinely "codes" a detailed specification into some computer language, a very denigrating designation for the work most computer programmers do.
In the 60s and 70s, with primitive tools, arcane computer languages and operating systems and generally little sophistication in the field, there may have been a place for separate designers and "coders". Nowadays, the designer (I'm talking about detailed design here, not about overall analysis and architecture) can easily "code" on the fly.
In other words anyone who can only code and not think and design as well has no place today.
Such extreme ripping only makes sense if you're listening on high-fi equipment. While the iPod has pretty decent quality for a portable player, it is impossible to hear the difference between 160kbit/s or better/higher with it (provided you use a decent encoder such as LAME).
Also, the higher bitrate wastes battery life proportionally, i.e. 256kbit/s shall use the battery twice as fast as 128kbit/s.
I use --alt-preset normal which is around 190kbit/s VBR, a good compromise between nearly as good as it gets for high-fi equipment and better than needed yet still not too big for portable use.
Slashdot Mirror
Why do ex-windows users keept claiming this? It is utter nonsense. X window has nothing to do with slow spread of unix on the desktop.
Do you really think business users are disturbed by too little FPS for 3 D shootere?!?
Yes, Motif is ugly. But it is not ugly because of X, but just because it is ugly. X is just a way to draw on a screen (locally or remotely). It is more than fast enough for any 2D needs one might have nowadays, even through a LAN. There are enough alternatives to Motif nowadays, but no single de facto standard has established itself yet. Do you think that by removing X and its network transparency, some elegant GUI toolkit suddenly establishes itself as de facto standard? Also on a direct FB it is possible to run many GUI toolkits.
There are more than enough hardware supported (2D) X windows drivers. How can you claim that X was designed for and ran well on a 8 Mhz 68000 CPU, and does not run well on current machines which are 100 times as fast?
As for people wanting desktop machines: "people" in fact want windows machines, because that is what they know and most are not interested in computers. However most incentives for introducing unix on the desktop in companies is to save cost, therefore the IT department/architecture forces the users whether they like it or not. Thus "thin clients" are introduced. Not the pure thin clients that are only a network terminal, but those that have the apps that are used most (office suite) installed locally and all other apps remotely, and running remotely via X.
That is a great solution, because it saves the IT department loads of local support and configuration. New version? No need to distribute software to 10000 desktops, just upgrade on a couple of servers. Why do you think even Microsoft has introduced "terminal server" after years of denial? Only when citrix got really popular they had to discover that many clients (i.e. large corporations) do want a mixture of desktop and thin clients.
Summary: it is an illusion and unfounded claim that X somehow prevents a migration to desktop unix. It is not because of, but rather in spite of X windows that migration to desktop unix is slow.
Especially in companies. At most places in companies where unix/linux is used, it is used by servers and thin clients. The network transparency is absolutely crucial. Even MSFT tried to add it to its GUI using terminal server.
Also at home it is useful to many. All the people I know that run linux or freebsd at home (except those that just try it out occasionally) have a linux or freebsd server and a windows or linux/freebsd desktop. Most apps are run through X, just 1 or 2 games that need more speed are run locally.
I think everyone how "grew up" on unix uses it this way. Only people that grew up on windows and just recently moved to linux are used to setting things up so that all interactive apps run directly on the desktop.
Which does not necessarily mean it shall be installed on 80%. In fact that number seems like a wild guess and very improbably to me.
If installing windows in vmware, be sure they have to pay the full price. In that case it would be impossible that the "linux" deal would be only a bit more expensive as the "windows" deal would have been. It should have cost double at least in that case.
However, it is well possible that they use a few servers with vmware GSX (the server variant) to which many PC's may occasionally connect via the vmware GSX client to run a legacy windows application. In that case a few windows licences would do.
What is buggy: the API or its implementation?
Strictly spoken, an API cannot be buggy. One could say it is inconvenient, badly structured, not well thought out or whatever.
But a bug is an implementation error, something does not work as specified. The API is the spec, only its implementation can be buggy.
What did you mean to say: Does the current NIO implementation contain bugs, or are the concepts of NIO broken?
Why does the top want XYZ? Because a customer (in the end a consumer) wants to buy it.
Who shall be left to buy, if all jobs except for the 3 top guys are "outsourced"?
Also: the consumer wants XYZ, he does not care if it is produced by a company in the US, Europe or India or China. So, the CEO of the US company shall loose his job at latest when the company ceases to exist, or is taken over by an Indian company.
There is nothing that makes the top guys more immune to this problem than others. It just will take a bit longer.
If Asian countries don't implement laws to effectively fight spam, this might be a way to force them. If people and businesses can no longer send us e-mail because personal servers, but also ISP's simply blacklist an entire country, I think it might have some effect.
Using relays based in other countries won't be possible either, since relays are also blacklisted.
you just block all mail coming from Asia.
90% of spam I received came from Asia (korea, china etc) until I added the following to my postfix access config file:
202 554 All sites from Asia-Pacific NIC blocked due to excessive SPAM
203 554 All sites from Asia-Pacific NIC blocked due to excessive SPAM
210 554 All sites from Asia-Pacific NIC blocked due to excessive SPAM
211 554 All sites from Asia-Pacific NIC blocked due to excessive SPAM
218 554 All sites from Asia-Pacific NIC blocked due to excessive SPAM
219 554 All sites from Asia-Pacific NIC blocked due to excessive SPAM
220 554 All sites from Asia-Pacific NIC blocked due to excessive SPAM
221 554 All sites from Asia-Pacific NIC blocked due to excessive SPAM
61 554 All sites from Asia-Pacific NIC blocked due to excessive SPAM
If people don't buy CD's long enough, the companies funding the RIAA shall go bankrupt and noone is left to fund the RIAA. Then the RIAA shall no longer have money to buy politicians and laws.
Because of this car makers have discouraged, even threatened, car dealers not to sell to foreigners, e.g. a danish dealer would risk his dealership if he sold to a german.
Now the EU has declared that illegal. The result is that the net prices (before VAT) shall be the same everywhere, so the price incl. VAT shall be much higher in denmark than in germany.
I think this is only reasonable. If some state decides it wants to have much higher taxes, the result is that the prices become much higher. It is unjust that consumers in countries with low taxes kind of subsidize those with high taxes.
So: what stops apple to offer their hardware in Hungary for the same net prices (excl. taxes and shipping) as in the US, then the consumer in hungary must pay extra for taxes and more expensive shipping. It should be up to him to decide, under those conditions, he still wants to buy.
Of course one cannot force a company to sell a product somewhere.
However I think in the case of webshops, I think a common reason is that in many countries there is still much more reliance on local dealers and importers; apple doesn't want to piss off its local dealers by competing with them by means of direct sales.
If I spread documents/e-books relying on such protection, I would not be a happy customer.
Sure they can sue makers of commercial cracking programs, but fact remains that many people can use such programs to make 'illegal' copies of my documents. Suing with the DMCA doesn't make much change in that respect.
Yes, MSFT shall not be able to keep up their rate of changing technology. They have been forced to do so and I think they shall suffer once they no longer can do this.
Reason: It has been their way of keeping competition shut out. As soon as others became able to reverse engineer their file formats, API's and network protocols, their products threatened to loose their exclusive nature and others would be able to create products that open a possible migration path. Once they (have to) stop doing this, it will become much easier for customers to move on, away from MSFT.
However, this only means that some (large) organisation has to be bold enough and take the first step, and say goodbye to MSFT office. And who is in a better position to do so (and force all parties that want to be involved with your organisations to adapt) than the state? It is the only organisation that can change laws and thus make this happen.
I expect that this strangling grip of MSFT Office can only be broken by governments/states to take a first step. This means that:
- citizens do not have to be forced to buy expensive software to deal with the state (which is immoral IMO)
- companies that want to deal with the state have to adapt and use an open file format as well
The benefits of being freed from the current hostage situation shall be enormous, and everyone shall be grateful for it, even if the initial steps may cause some inconvenience and money. Sometimes you have to be not so shortsighted plan a bit further ahead.Ion is a descendant of PWM that is being actively maintained. I've been using it since the last year (after 15 years of twm and ctwm).
evilwm, I would say, is a close second. It happens to have exactly the bindings that I configured for twm. But I do like the idea of PWM and ION to always maximize windows and abolish the overlapping windows concept. It makes maximum use of your screen and minimizes the need to use a mouse for window management tasks.
Generally, I'm sceptic towards syntactic sugar (such as operator overloading, or even overloading in general). However, if you call generics syntactic sugar, you don't know what they are or what you are taling about.
Sure, its implementation may be in terms of what already exists (i.e. internally you still have Lists of Objects, being cast into Integers or whatever). However, the burden of typecasting is removed from the programmer into the compiler, which is very significant and not syntactic sugar at all!
C++ in its first incarnations was also just a precompiler to C, so following your reasoning one might have called C++ just syntactic sugar. Or even a compiler as mere syntactic sugar w.r.t. assembly, since in the end it is machine code that runs in both cases.
No, getting rid of typecasts, a source of many errors and mistakes, is a big step towards a more safe and better readable language.
As far as the remaining lack of stack variables: I agree it would be nice to have them. However I consider this an implementation/optimization detail. It should be up to the compiler to decide whether to use stack or heap for a particular case (though I'm not sure if current compilers can make the best decision). Kind of like the 'register' keyword in C, that hinted the compiler to store an important variable in a register. Due to progress in compiler technology it has become totally redundant and even harmful.
But don't forget that even thoug landline phone is relatively expensive, mobile is even more expensive. Especially if using it across national borders, it is obscenely expensive.
Also for data (modem, ADSL) mobile phone is not really an alternative.
I see mobile telephony mostly as fashionable, but less functional (slow data speed, very expensive).
It is a programming language (stack based, like forth), which is amazing for its primary purpose: printing.
But to call the language itself "extremely powerful" is an exaggeration. As a programming language it is quite primitive and incomprehensible, compared to more powerful languages such as C++, java, ML, yes even forth.
There is a big difference between SUN and SGI (apart from what others have said here):
The E15K is used by big companies such as banks as mainframe like or mainframe replacement computers.
The fast SGI machines were mainly used as scientific supercomputers, which is a small niche market.
The mainframe market for superservers and UNIX consolidation however is huge and no niche market at all!
Where I work (2nd bank of Switzerland) it is E15K's all the way. All software written in house (by thousands of developers) runs either on MVS or on E15K, partitioned to serparate and guarantee applications the capacity they need.
Apart from the fact that X is very good, it is absolutely crucial that Linux remains compatible with the rest of the "free world", that is UNIX and *BSD.
Remember, these are all allies against a common enemy. Going it alone in a crucial such as the GUI would be fatal for us all, and that further fragmentation be avoided.
Read/(write) access to none native filesystems such as NTFS is only relevant for dual boot systems, where it may be nice to access your NTFS partition while booted into *BSD or Linux.
But "real" systems are no dual boot systems. So you don't need it. Hardly find NTFS on floppies or CD-R or on tape. OK maybe for a hotswappable (scsi) harddisk it might have a use, but that is the only serious thing I can think of.
All other interoperability between filesystems goes via network filesystems, be it SMB, NFS, AFS, DFS or whatever. Those are the ones you need. A good and free implementation of NFS for NT might be nice, or Samba keeping up with ever changing Windows fileservers. For real use you need a directory service across platforms as well (NIS+ on windows, active directory on UNIX).
It may not allow the ad-hoc fine-tuning that ACL's allow, but IMO that is a good thing.
Really, ACL's may quickly become a mess. Yes you can set some special permissions and exceptions fast, without too much planning or thinking (as groups require) but after a while it is unmanageable. Why do you think that many larger sites running large Windows implementations have a policy of not allowing ACL?!?
But still, a few years ago you'd have to buy the disc again in such cases. Nowadays you just burn again from the original. 100% legitimate of course, but it is costing them turnover compared to the old days.
I can only hope that their profits shrink so fast that they cannot afford to buy laws much longer.
I hardly see any place left for "coders" these days. The word "coder" implies someone who routinely "codes" a detailed specification into some computer language, a very denigrating designation for the work most computer programmers do.
In the 60s and 70s, with primitive tools, arcane computer languages and operating systems and generally little sophistication in the field, there may have been a place for separate designers and "coders". Nowadays, the designer (I'm talking about detailed design here, not about overall analysis and architecture) can easily "code" on the fly.
In other words anyone who can only code and not think and design as well has no place today.
I always get the creeps when I see if( xxx ).
It is as if "if" is a function putting the parentheses like that.
foo( xxx ) is a function call
if ( xxx ) is a statement which gets a block in parentheses.
I couldn't resist. I have to fight this good fight every day against some of my colleages.
Such extreme ripping only makes sense if you're listening on high-fi equipment. While the iPod has pretty decent quality for a portable player, it is impossible to hear the difference between 160kbit/s or better/higher with it (provided you use a decent encoder such as LAME).
Also, the higher bitrate wastes battery life proportionally, i.e. 256kbit/s shall use the battery twice as fast as 128kbit/s.
I use --alt-preset normal which is around 190kbit/s VBR, a good compromise between nearly as good as it gets for high-fi equipment and better than needed yet still not too big for portable use.
Ugh, it does not have an FM tuner? That is incredible. I almost planned to buy an iPod not even thinking it might not have an FM tuner.
And the Zen is too big & heavy. I think I'll stick with my RAM based player (with FM tuner) for a while then.
Have people been bugging Apple about this omission?