With all due respect, Monty's article came across as advanced whining. This is no language issue.
Erm - the title of this thread is "That Widenius guy should learn English" - that sounds like a language issue, although I take the point that he may be whining like a 5 year old.
In fact, I think he is saying exactly what we might expect him to say, under the circumstances. The Postgres people posted a silly, one sided press release based on some undisclosed figures. Predictably, the MySQL people have now posted a response, and actually have some figures on their website. We all carry on, safe in the knowledge that benchmarks tell you nothing:-)
And of course, in the real world, nobody ever actually inserts and selects at the same time.
There are a fair number of applications, especially web-based ones, where selects will be far more frequent than inserts. It depends entirely on the application.
This is one of the attitudes I don't understand. Why shouldn't I have every blasted service turned on if I want to? Are the Linux services so riddled with security holes that the only safe configuration is to have the whole computer turned off?
No, but holes do get found in services on all operating systems. If you are only running services that you have explicitly turned on you will seethe advisories and upgrade. If you have all services turned on by default, users won't be aware that they are running services, let alone what they are and will not upgrade when required.
You only need to be running one service with a hole in it to get compromised. Having Linux machines compromised through services that their owners didn't know they were running gains Linux a very poor reputation for security: Linux suffered greatly in our university when passwords were sniffed using a Linux box compromised via the famous imapd exploit - an imapd that the owner didn't know they were running. As a result we see college policies banning the connection of Linux boxes to the network.
If you want to run "every blasted service" then by all means go and turn them all on. It is then up to you to keep abreast of all security updates. This is a very different situation from having all the services turned on by default.
Bottom line: The RBL is evil because it limits my connectivity to the outside world - and it does so in a transparent manner in its typical configuration. I do not want my local net admin/nazi to control *MY* e-mail - maybe I WANT e-mail from cyberpromotions, maybe not. But it is MY choice, not YOURS. And that's where I draw the line on the "free speech" of this system.
This is your choice, so go pay for an account with an ISP that doesn't subscribe to the RBL. People pay good money for ISPs that *do* subscribe, because it saves them from wasting money downloading spam (yes, in some parts of the world people still pay for their time online). If people object to this form of censorship, they can go buy an RBL-free account. Most people are perfectly happy with MAPS' judgement on this.
MAPS are only as powerful as their subscribers make them. Any administrator signing up to MAPS RBL is saying, on behalf of all of their users, that they are happy to not recieve mail coming from any server listed on the RBL.
MAPS are not forcing this on anyone. People sign up to MAPS because we trust their judgement on what is and isn't mail abuse. If they start turning up too many false positives, people will unsubscribe from them as the number of complaints from users that can't recieve email from people they want to starts to exceed the number of complaints about spam. For example, many people avoid signing up to ORBS because they find their policy too cavalier. It's a self regulating system.
By signing up to the RBL, people are effectively saying "we don't want to recieve mail from you unless you comply with MAPS' policy. Deal with it."
Somebody has already mentioned not to confuse the opinions of the author with the opinions of KDE authors, but you have a point.
What is different here is that the developer has been brought to the project from a non-open source background. Open source developers have previously come to write stuff because they have an itch to scratch - they want to help develop the software so much that they are happy to do it for nothing, in their spare time.
There is bound to be some conflict when you bring people to a project who are there because they are paid to be rather than because they really want to be there, although sometimes the two will happily coincide.
Open source projects are usually run as a meritocracy, so if the people that Corel is paying to work on this are not doing anything useful, their code won't make it into the main CVS tree. This is one of the great things about open source development - qualifications and experience count for little compared to the ability to produce quality code.
eading your requirements, you should definitely take a look at how the Abiword people approach this.
This project is going well in that they have succesfully attracted people from the community into making serious contributions, as well as a core group of employees who spend a varying amount of time on the project. If your project could reach this stage then it would make sense for your company financially as some of the development effort would be coming from volunteers in the community.
I think that "the community" is naturally suspicious of companies going the open source route -- are they genuine or are they just jumping on the bandwagon? -- and if you expect serious contributions you'll need to do it on the community's terms: Go with the GPL, or some other well known license.
The way that the AbiWord people approach this is that the code is GPL but they (the company) retain the trademark so that only their releases may be called AbiWord.
As for who gets to submit code: run it as a meritocracy. Initially only your employees have access, and others may only submit patches. Regular submitters of quality patches should then be considered for full CVS access.
Perhaps they shuld use a truly private directory on the mirrows, then mv it across on the third stroke.
It's a nice idea, but it also requires extra effort on the mirror admin's part. Most of the sites will be mirroring more than just Mandrake and don't have the time to implement this for every package mirrored.
It doesn't take a rocket scientist to come to the conclusion that in order to give "priority" to a site, then something must be taken away from another.
Not necessarily. The extra money from these sites will allow them to improve their connectivity to other networks, and they make this extra bandwidth available preferentially to sites that have funded it.
You're assuming that the one and only bottleneck is at the modem/xDSL end, which is not the case.
Slashdot Mirror
With all due respect, Monty's article came across as advanced whining. This is no language issue.
:-)
Erm - the title of this thread is "That Widenius guy should learn English" - that sounds like a language issue, although I take the point that he may be whining like a 5 year old.
In fact, I think he is saying exactly what we might expect him to say, under the circumstances. The Postgres people posted a silly, one sided press release based on some undisclosed figures. Predictably, the MySQL people have now posted a response, and actually have some figures on their website. We all carry on, safe in the knowledge that benchmarks tell you nothing
That article looks like it was written by a 5-year-old.
Or indeed someone whose first language is not English. Someone from Finland for example.
This man clearly has taken the time to learn English - could you have written that article in Finish? Or indeed any other language other than English?
And of course, in the real world, nobody ever actually inserts and selects at the same time.
There are a fair number of applications, especially web-based ones, where selects will be far more frequent than inserts. It depends entirely on the application.
This is one of the attitudes I don't understand. Why shouldn't I have every blasted service turned on if I want to? Are the Linux services so riddled with security holes that the only safe configuration is to have the whole computer turned off?
No, but holes do get found in services on all operating systems. If you are only running services that you have explicitly turned on you will seethe advisories and upgrade. If you have all services turned on by default, users won't be aware that they are running services, let alone what they are and will not upgrade when required.
You only need to be running one service with a hole in it to get compromised. Having Linux machines compromised through services that their owners didn't know they were running gains Linux a very poor reputation for security: Linux suffered greatly in our university when passwords were sniffed using a Linux box compromised via the famous imapd exploit - an imapd that the owner didn't know they were running. As a result we see college policies banning the connection of Linux boxes to the network.
If you want to run "every blasted service" then by all means go and turn them all on. It is then up to you to keep abreast of all security updates. This is a very different situation from having all the services turned on by default.
Bottom line: The RBL is evil because it limits my connectivity to the outside world - and it does so in a transparent manner in its typical configuration. I do not want my local net admin/nazi to control *MY* e-mail - maybe I WANT e-mail from cyberpromotions, maybe not. But it is MY choice, not YOURS. And that's where I draw the line on the "free speech" of this system.
This is your choice, so go pay for an account with an ISP that doesn't subscribe to the RBL. People pay good money for ISPs that *do* subscribe, because it saves them from wasting money downloading spam (yes, in some parts of the world people still pay for their time online). If people object to this form of censorship, they can go buy an RBL-free account. Most people are perfectly happy with MAPS' judgement on this.
MAPS are only as powerful as their subscribers make them. Any administrator signing up to MAPS RBL is saying, on behalf of all of their users, that they are happy to not recieve mail coming from any server listed on the RBL.
MAPS are not forcing this on anyone. People sign up to MAPS because we trust their judgement on what is and isn't mail abuse. If they start turning up too many false positives, people will unsubscribe from them as the number of complaints from users that can't recieve email from people they want to starts to exceed the number of complaints about spam. For example, many people avoid signing up to ORBS because they find their policy too cavalier. It's a self regulating system.
By signing up to the RBL, people are effectively saying "we don't want to recieve mail from you unless you comply with MAPS' policy. Deal with it."
Somebody has already mentioned not to confuse the opinions of the author with the opinions of KDE authors, but you have a point.
What is different here is that the developer has been brought to the project from a non-open source background. Open source developers have previously come to write stuff because they have an itch to scratch - they want to help develop the software so much that they are happy to do it for nothing, in their spare time.
There is bound to be some conflict when you bring people to a project who are there because they are paid to be rather than because they really want to be there, although sometimes the two will happily coincide.
Open source projects are usually run as a meritocracy, so if the people that Corel is paying to work on this are not doing anything useful, their code won't make it into the main CVS tree. This is one of the great things about open source development - qualifications and experience count for little compared to the ability to produce quality code.
eading your requirements, you should definitely take a look at how the Abiword people approach this.
This project is going well in that they have succesfully attracted people from the community into making serious contributions, as well as a core group of employees who spend a varying amount of time on the project. If your project could reach this stage then it would make sense for your company financially as some of the development effort would be coming from volunteers in the
community.
I think that "the community" is naturally suspicious of companies going the open source route -- are they genuine or are they just jumping on the bandwagon? -- and if you expect serious contributions you'll need to do it on the community's terms: Go with the GPL, or some other well known license.
The way that the AbiWord people approach this is that the code is GPL but they (the company) retain the trademark so that only their releases may be called AbiWord.
As for who gets to submit code: run it as a meritocracy. Initially only your employees have access, and others may only submit patches. Regular submitters of quality patches should then be considered for full CVS access.
Perhaps they shuld use a truly private directory on the mirrows, then mv it across on the third stroke.
It's a nice idea, but it also requires extra effort on the mirror admin's part. Most of the sites will be mirroring more than just Mandrake and don't have the time to implement this for every package mirrored.
It doesn't take a rocket scientist to come to the conclusion that in order to give "priority" to a site, then something must be taken away from another.
Not necessarily. The extra money from these sites will allow them to improve their connectivity to other networks, and they make this extra bandwidth available preferentially to sites that have funded it.
You're assuming that the one and only bottleneck is at the modem/xDSL end, which is not the case.