Apache 2 was faster than Apache 1.3 for me in general, so the answer is "probably yes".
However, I consider (Apache|ModPerl)::Registry.* harmfull in production environment. It makes you think you can get away with quick CGI-script coding and may hit unexpected problems on your way. So if you decide to switch to Apache 2 / mod_perl 2, why not rewrite the thing the Apache::Request(2?) way?
It really depends on the way you coded your CGI application. If you use CGI.pm, you probably load and have loaded quite a lot of bloat (despite the CGI.pm's AUTOLOADing features). Apache::Request is C library, so it is marginally faster than a script using CGI.pm.
Also, once you rewrite your application the Apache::Request way, you'll probably refactor it to be cleaner, leading to faster executions. Coding with mod_perl in mind lead to better code and faster applications for me. Since you know that your handlers will be executed multiple times, you cannot get away with fast hacking like you do with pure CGI.
Usually the information is available from multiple sources. So if you do not need the author of premium newspaper to write the article for you, you can always get the core of the story elsewhere.
That is why when the newspapers make distinction, they make it the other way round -- the fresh info is free and you have to register or pay for their back stories and archives.
The trouble is that most of the spam comes from compromised PCs nowadays. So the person does not need to give your address to anybody -- you will start getting spam once virus finds its way to their system. I even receive spam to an address I gave to employees of very reputable company (where I'm sure they did not give out the info to marketeers).
returning PNG. So yes, it works by using processing on toolbar.nickstallman.net and you can easily call the functionality from any browser you like, however, it will be easy for Google to close access for him when they see fit, so it does not look like a long-term solution.
On the other hand, this was the first time that I saw PageRank for some of my pages, so it is a nice toy.
It thinks it handles UTF-8 (puts it into the generated HTML pages) but the editting pages run in ISO-8859-1, making it nearly impossible to enter ISO-8859-x characters for x != 1. And the pages lack email address of the maintainers of the Wiki. Does anyone know how swpatwiki.ffii.org could be fixed to run in UTF-8 fully?
You have obviously no experience with an Oracle database?
You mean besides doing DBA and PL/SQL development on a daily basis?
First of all, Oracle software is not installed as 'root', but as 'oracle' and as a memeber of 'oinstall' and 'dba' groups.
Second, Oracle software can be installed as any user and group you please (except root, which I believe is still technically possible, even if not recommended). Third, rpm can very well install the software under any user and group it chooses, the same way it does with mailservers, HTTP server or database servers. The same holds for linking anything to anything else and setting up default configuration. Besides, rpm could easily handle dependencies.
i will repeat again: Oracle is not a click and run isntallation, but requires knowledge other than just typing install or rpm -Uvh. until you understand that, you will continue to tout rpm based installs, but that only shows an Oracle savvy person that you have no clue whatsoever!
Your claims are being contradicted by Oracle Corp. Did you install Oracle Database lately? They made it into./runInstaller, click, click, click, OK, click, click some more, and run thing, as far as setting the software fast is concerned. You do not need to be fluent with names of v$ and dba_ views to run the database with OEM, 10g has ADDM, the SGA can be sized by the database server,... Yes, as I've said, for production environment you better know what you are doing, but that is by no means unique to Oracle. Oracle is a piece of software. It is not a shame to make software easily usable
By minor problems I mean having to set the LD_ASSUME_KERNEL environment variable, the problem of ins_ctx.mk, or having to touch/etc/rac_on even if you do not run RAC.
Having installed and upgraded Oracle on multiple RH and Fedora boxes, I did not get the "plain doesn't work" result.;-)
I am eagerly waiting for PostgreSQL to have reasonable exception handling in procedural languages (PL/pgSQL). Without this you can hardly do heavy-duty server-side programming.
In addition, Oracle making Linux its primary development and target platform means that there should be less obstacles actually using the (Oracle) software on Linux machines. If companies get new versions and patches for Linux earlier than for Windows versions of the database software, it might make the swich decission easier.
Actually, non-enterprise versions of Red Hat, as well as Fedora, will do just fine. There might be some minor problems, but Google is your friend in resolving them (or ignoring bogus error messages). After you install the software, there is no problem running it in a rock-stable fashion.
More like a reliable SUV, I'd say. With the multithreading servers, the overhead of database connection drops significantly, and it can be further tuned down with mod_perl persistent connections.
The biggest pain I've had with Oracle lately is their focus on graphical-only installation and configuration tools, with Java versions being rather critical factor of success.
Whether the database is disk intensive depends heavily on the type of applications you run on top of the database. For many Web applications, most of the operations is reading and you can have most of the data you need cached in RAM. The throughput of the database system is also heavily dependant on the way you use (or abuse) the database and its transaction manager.
As for the licencing fees, according to oraclestore.oracle.com, Oracle Standard Edition One costs USD 999 per processor per year. It is perfectly possible to run Oracle database on stock PC hardware, making it possible to upgrade to new processor and bus speeds at a fraction of cost of more up-scale hardware.
If the spammers get as much virus-bounces as I do (about 1000 over last 24 hours), they figure out that it doesn't make sense to send any spam at the moment. It will simply get lost among the other trash.
I set up bogofilter to mark Bogosity in two categories -- viruses and spam. Then I color the index in my mutt accordingly and I get nice overview. The virus to spam ratio is about 25 : 1. The spam to legal mail ratio is about 3 : 1.
Well, I couldn't resist joking a little. Hope no one is offended too much.
I believe that either he is looking for an answer to his immediate need, and then he should be as specific as possible. At least saying how much he expect to travel and how much he expects to use the plan would be nice. And on which cost he is based. Or the question was more like "I'd be interested to know what are the GPRS conditions around in U.S., Canada, Europe and Asia, to get a broad picture". Then he should have said so and I'd happily bring in some European quotes.
You are correct and I have no problem with Slashdot being U.S.-centric (certainly wouldn't mind otherwise, but can live with it:-). However, the original post was nonsense even in U.S.-centric media, IMHO. He failed to say where in the U.S. he were, where he travels, what are his usage patterns... You cannot give any useful help because on the other coast, conditions may be completely different.
To do the translation, you don't need any permission. You can translate for your own use ad infinum.
Now, if you bought the book in English (and paid royalties for the "ideas" in the book), should you pay for those "ideas" again if you buy the book again, translated into German? Or should you get some discount and only pay for the "translation" part?
The same with music: if I once bought the song on LP (so I got a right to use it, to listen to it), should I pay again _for the song_ when I buy it on CD? Or should I only pay for the medium a technicalities?
Please note that robots.txt affects whether Google crawls various parts of your website at all. To prevent your pages from being stored in the Google cache (even if they are searchable using Google), you need to specify the META tag
<META NAME="GOOGLEBOT" CONTENT="NOARCHIVE"> in each and every of your pages.
Slashdot Mirror
Apache 2 was faster than Apache 1.3 for me in general, so the answer is "probably yes".
However, I consider (Apache|ModPerl)::Registry.* harmfull in production environment. It makes you think you can get away with quick CGI-script coding and may hit unexpected problems on your way. So if you decide to switch to Apache 2 / mod_perl 2, why not rewrite the thing the Apache::Request(2?) way?
It really depends on the way you coded your CGI application. If you use CGI.pm, you probably load and have loaded quite a lot of bloat (despite the CGI.pm's AUTOLOADing features). Apache::Request is C library, so it is marginally faster than a script using CGI.pm.
Also, once you rewrite your application the Apache::Request way, you'll probably refactor it to be cleaner, leading to faster executions. Coding with mod_perl in mind lead to better code and faster applications for me. Since you know that your handlers will be executed multiple times, you cannot get away with fast hacking like you do with pure CGI.
Usually the information is available from multiple sources. So if you do not need the author of premium newspaper to write the article for you, you can always get the core of the story elsewhere.
That is why when the newspapers make distinction, they make it the other way round -- the fresh info is free and you have to register or pay for their back stories and archives.
The trouble is that most of the spam comes from compromised PCs nowadays. So the person does not need to give your address to anybody -- you will start getting spam once virus finds its way to their system. I even receive spam to an address I gave to employees of very reputable company (where I'm sure they did not give out the info to marketeers).
True but NY Times.com will present you with colorful ads when you read it on-line.
A quick grep through the source shows that the functionality of the plugin can be summarized as
= ht tp://www.asdf.com/
http://toolbar.nickstallman.net/toolbar.php?url
returning PNG. So yes, it works by using processing on toolbar.nickstallman.net and you can easily call the functionality from any browser you like, however, it will be easy for Google to close access for him when they see fit, so it does not look like a long-term solution.
On the other hand, this was the first time that I saw PageRank for some of my pages, so it is a nice toy.
It thinks it handles UTF-8 (puts it into the generated HTML pages) but the editting pages run in ISO-8859-1, making it nearly impossible to enter ISO-8859-x characters for x != 1. And the pages lack email address of the maintainers of the Wiki. Does anyone know how swpatwiki.ffii.org could be fixed to run in UTF-8 fully?
You mean besides doing DBA and PL/SQL development on a daily basis?
Second, Oracle software can be installed as any user and group you please (except root, which I believe is still technically possible, even if not recommended). Third, rpm can very well install the software under any user and group it chooses, the same way it does with mailservers, HTTP server or database servers. The same holds for linking anything to anything else and setting up default configuration. Besides, rpm could easily handle dependencies.
Your claims are being contradicted by Oracle Corp. Did you install Oracle Database lately? They made it into ./runInstaller, click, click, click, OK, click, click some more, and run thing, as far as setting the software fast is concerned. You do not need to be fluent with names of v$ and dba_ views to run the database with OEM, 10g has ADDM, the SGA can be sized by the database server, ... Yes, as I've said, for production environment you better know what you are doing, but that is by no means unique to Oracle. Oracle is a piece of software. It is not a shame to make software easily usable
By minor problems I mean having to set the LD_ASSUME_KERNEL environment variable, the problem of ins_ctx.mk, or having to touch /etc/rac_on even if you do not run RAC.
;-)
Having installed and upgraded Oracle on multiple RH and Fedora boxes, I did not get the "plain doesn't work" result.
You might be right for production environment. There you would well expect the DBA to know what they are doing.
However, having the installation in the form of
rpm -Uvh oracle-standard-one-10.1.0.2.i386.rpm
would certainly be nice. After all, it is just software.
I am eagerly waiting for PostgreSQL to have reasonable exception handling in procedural languages (PL/pgSQL). Without this you can hardly do heavy-duty server-side programming.
In addition, Oracle making Linux its primary development and target platform means that there should be less obstacles actually using the (Oracle) software on Linux machines. If companies get new versions and patches for Linux earlier than for Windows versions of the database software, it might make the swich decission easier.
Actually, non-enterprise versions of Red Hat, as well as Fedora, will do just fine. There might be some minor problems, but Google is your friend in resolving them (or ignoring bogus error messages). After you install the software, there is no problem running it in a rock-stable fashion.
More like a reliable SUV, I'd say. With the multithreading servers, the overhead of database connection drops significantly, and it can be further tuned down with mod_perl persistent connections.
The biggest pain I've had with Oracle lately is their focus on graphical-only installation and configuration tools, with Java versions being rather critical factor of success.
Whether the database is disk intensive depends heavily on the type of applications you run on top of the database. For many Web applications, most of the operations is reading and you can have most of the data you need cached in RAM. The throughput of the database system is also heavily dependant on the way you use (or abuse) the database and its transaction manager.
As for the licencing fees, according to oraclestore.oracle.com, Oracle Standard Edition One costs USD 999 per processor per year. It is perfectly possible to run Oracle database on stock PC hardware, making it possible to upgrade to new processor and bus speeds at a fraction of cost of more up-scale hardware.
You have just described SPF, see http://spf.pobox.com/
If the spammers get as much virus-bounces as I do (about 1000 over last 24 hours), they figure out that it doesn't make sense to send any spam at the moment. It will simply get lost among the other trash.
I set up bogofilter to mark Bogosity in two categories -- viruses and spam. Then I color the index in my mutt accordingly and I get nice overview. The virus to spam ratio is about 25 : 1. The spam to legal mail ratio is about 3 : 1.
Well, I couldn't resist joking a little. Hope no one is offended too much.
I believe that either he is looking for an answer to his immediate need, and then he should be as specific as possible. At least saying how much he expect to travel and how much he expects to use the plan would be nice. And on which cost he is based. Or the question was more like "I'd be interested to know what are the GPRS conditions around in U.S., Canada, Europe and Asia, to get a broad picture". Then he should have said so and I'd happily bring in some European quotes.
cdma phones dont have sim cards which is certainly a very positive feature.
You mean that SIM cards are positive feature or that not having them is a positive feature? If the later, could you please elaborate?
You are correct and I have no problem with Slashdot being U.S.-centric (certainly wouldn't mind otherwise, but can live with it :-). However, the original post was nonsense even in U.S.-centric media, IMHO. He failed to say where in the U.S. he were, where he travels, what are his usage patterns ... You cannot give any useful help because on the other coast, conditions may be completely different.
True. But releasing a product that doesn't have GSM/GPRS, Bluetooth nor WiFi? Do people still like to fiddle with cables?
Most any European operator will provide you with reasonable GPRS plan.
Oh wait, you're not in Europe but in Singapore? Or Angola? Well, then bad luck, I have no idea, you should have stated _where_ you were.
Administrators, could we just skip this kind of submissions that somehow forget that there's world behind borders on poster's Iowa house?
Why move anywhere? It's perfect.
To do the translation, you don't need any permission. You can translate for your own use ad infinum.
Now, if you bought the book in English (and paid royalties for the "ideas" in the book), should you pay for those "ideas" again if you buy the book again, translated into German? Or should you get some discount and only pay for the "translation" part?
The same with music: if I once bought the song on LP (so I got a right to use it, to listen to it), should I pay again _for the song_ when I buy it on CD? Or should I only pay for the medium a technicalities?
Please note that robots.txt affects whether Google crawls various parts of your website at all. To prevent your pages from being stored in the Google cache (even if they are searchable using Google), you need to specify the META tag <META NAME="GOOGLEBOT" CONTENT="NOARCHIVE"> in each and every of your pages.