3 ms·

Just as a warning: Manuel Lemos is the most clueless and destructive clown in the entire PHP community. PHPclasses is a collection of worst practices and anti-p

Just as a warning: Manuel Lemos is the most clueless and destructive clown in the entire PHP community. PHPclasses is a collection of worst practices and anti-patterns. Nothing he writes is in any way plausible, it's all just a feeble attempt at arguing that any engineering practice he can't wrap his head around is wrong. The man is just desperately trying to defend his limited understanding of software development. Excuse the ad hominem, but there's really zero plausible content to this, or any other of the delusional utterings he has published over the last decade.

h2s 13y ago
I agree in as much as I think that PHPClasses is an anachronistic throwback to the bad old days of PHP. It has a shameful history of promoting terrible code with dubious "awards" as if it were the greatest thing ever built. Newcomers to PHP development would be better served by a web that lacked phpclasses.org in its current form. For example, in a recent article entitled "26 Ways to Show that PHP Can Be Better Than PHP" (http://www.phpclasses.org/blog/post/203-26-Ways-to-Show-that-PHP-Can-Be-Better-Than-PHP.html http://www.phpclasses.org/blog/post/203-26-Ways-to-Show-that...), they showcased a series of "multithreading" implementations in ascending order of WTF, culminating with this: Mohammed Yousef from Egypt has written the PHP Threader class that also takes advantage of the fact that a Web server can run multiple scripts in parallel, except that it uses AJAX requests sent from the browser to start the scripts. I disagree, however, with the idea that Manuel Lemos is a clueless, destructive clown. I think he's a well-meaning, intelligent person who has somewhat failed to keep pace with the "proper software engineering" standards that have infiltrated PHP in recent years. For example, in an article from 2013 comparing development using Wordpress vs using a framework, I would have expected some comparison of the viability of automated testing in each approach. And unless my name was Fabien Potencier, I would not expect to be advised to "use your own framework if you have a good capable framework to address your needs". But this is only destructive to the extent that others fail to make a strong case for building well-tested software using high quality open source libraries where suitable instead of endlessly reinventing wheels. There is room for competing ideas, and I don't see these particular ideas as an especially worrying competitive threat.

h2s 13y ago

I agree in as much as I think that PHPClasses is an anachronistic throwback to the bad old days of PHP. It has a shameful history of promoting terrible code with dubious "awards" as if it were the greatest thing ever built. Newcomers to PHP development would be better served by a web that lacked phpclasses.org in its current form. For example, in a recent article entitled "26 Ways to Show that PHP Can Be Better Than PHP" (http://www.phpclasses.org/blog/post/203-26-Ways-to-Show-that-PHP-Can-Be-Better-Than-PHP.html http://www.phpclasses.org/blog/post/203-26-Ways-to-Show-that...), they showcased a series of "multithreading" implementations in ascending order of WTF, culminating with this: Mohammed Yousef from Egypt has written the PHP Threader class that also takes advantage of the fact that a Web server can run multiple scripts in parallel, except that it uses AJAX requests sent from the browser to start the scripts. I disagree, however, with the idea that Manuel Lemos is a clueless, destructive clown. I think he's a well-meaning, intelligent person who has somewhat failed to keep pace with the "proper software engineering" standards that have infiltrated PHP in recent years. For example, in an article from 2013 comparing development using Wordpress vs using a framework, I would have expected some comparison of the viability of automated testing in each approach. And unless my name was Fabien Potencier, I would not expect to be advised to "use your own framework if you have a good capable framework to address your needs". But this is only destructive to the extent that others fail to make a strong case for building well-tested software using high quality open source libraries where suitable instead of endlessly reinventing wheels. There is room for competing ideas, and I don't see these particular ideas as an especially worrying competitive threat.