On Sunday, 09 September 2007, at 23:49:47 (-0400), Youness Alaoui wrote: > First, before I answer each of your statements below, I'd like to > say that your email was rather rude, and it looks like you are > biaised. I'm afraid that your answer would open a troll, which is > not what I want.
My e-mail was not rude. What is rude is to presume to tell the authors of other projects what they should do with their own work. And everyone is biased. Anyone who has given the issue enough logical, considered thought to form an opinion on the situation is "biased," and anyone who has not has no business giving an opinion to begin with. I presented the situation as I see it, fairly and rationally. What's important is not that I have an opinion, but rather than I have no vested interest in either project. I am not an author of either library nor of any project that depends on or is driven by either library. > The activity doesn't represent much either. CodeWarrior has been > busy lately, so it could be the reason for his inactivity. There's the keyword: inactivity. The reason is immaterial: I referred only to activity or the lack thereof. The author(s) being "otherwise occupied" is the very definition of "inactive." (See also: Eterm. Just because I have very good reasons for not being able to work on it lately doesn't make it any more "active.") > Also, CVS commit counts doesn't represent the real status of a > project, it actually depends on what they provide and the overall > stability. Clearly that doesn't stop people from telling you it's a reason to use one library over another, since that's exactly what you claim they did. > I'm not talking here about what the human eye/brain can > perceive. But if you want more details. I was refering to the wiki > page about EWL performance. That's one of the main arguments used by > the EWL team, and it is true that EWL looks MUCH more performant if > we reason by looking at those numbers. I know exactly what you were talking about. I saw the numbers when they were first posted. > But seriously, how many applications use 100K widgets ? That's a silly question. How many applications sort millions of pieces of data? Any that want to. How many applications use 100,000 widgets? Any that want to. > What are the "real" performance numbers ? There's no such thing. Even beginning computer science courses cover the basic futility of benchmarking, the need for mathematical representations of efficiency and complexity (like Big O notation), etc. I shouldn't need to sit here and explain that. > If we compare each and every widget provided by EWL and ETK, which > ones are faster? what if we use 10 widgets and not 100K widgets? > which toolkit is more performant ? EWL. > That test was merely a scalability test, that's all it is. That's simply false. Because computers are so fast, we often have to use large, even unrealistic quantities of data to generate numerically significant differences in measurements when comparing algorithms and implementations. Run "make perf" in libast sometime; some routines have to run through millions of iterations just to get numerically significant runtimes. Scalability indicates efficiency. Single-pass performance tests are not statistically valid because there are too many outside variables. But when we scale things up, those variations become less significant and the results more valid. It's the same principle. If EWL is more efficient with 100,000 widgets, it's because it's more efficient with 1 or 2 widgets. > I don't plan on use ETK or EWL in such proportions. If I needed to > create a chart with hundreds or thousands of buttons and labels then > ok, EWL is better, but for a "normal" application, I don't know > which one is more performant that the other, Of course you do. > and even if EWL is still more performant, if it's a 5% increase in > performance, is it worth it to choose EWL over ETK > because of performance, if overall, EWL responds less to our demands... What does that mean? No one has said anything about "responding to demands" until now. That doesn't even make any sense in the context of this conversation. > That's what I'm talking about, so please, no need to use such an argument. That's the problem. You don't seem to know what you're talking about, or at least about the issues surrounding it. No offense intended; you just don't seem to grasp the concepts at hand. > And by the way, quicksort might be faster than bubble sort, but only > for large numbers of 'n'. I'm sure that bubble sort is faster on > smaller samples. This statement seems to illustrate my point quite effectively. You could not be more mistaken. > When I said equivalent, I meant from an end user point of view, or > even from a developer point of view, using this toolkit or the other > is the same to you. And I'm saying you're wrong. :) > Don't take it as a personal attack because I compared EWL with ETK. I have nothing personal invested in it. I don't take anything personally, nor should anyone else. > Why such agressivity when you're talking to me? The word is "aggression." :) And I'm not aggressive. I'm blunt. There's a big difference. And I'm blunt with everyone, not just you. If you find logic, conviction, and proper diction "aggressive," there's really nothing I can do about that. That's how I talk. Or at least write. ;-) > Don't take my sentence out of its context. I never asked anyone to > "do my homework" or make decisions for me, or whatever you want to > call it. The following sentence is part of this paragraph you just > quoted. I'm not asking for one toolkit, I'm not asking raster to > make a choice and tell everyone to use this toolkit and not the > other, but I'm asking for a non-biaised opinion of someone not > belonging to either team to state what are the pros and cons of each > one, and which one is more suited for which type of use. I took nothing out of context. That's why I always quote the original e-mail as completely as possible -- so that there's no chance of anything being construed as "out of context." I keep the context right there! You asked for an "official" proclamation. Only raster can give such a thing. And as I said, there won't be one. > Because I'm asking for a point of view coming from someone not > belonging to either one of the teams. And you just received it. I don't belong to either team. > > > Right now, I'm working with ETK *ONLY* because cmarcelo is writing > > > ETK bindings for python and I need to work with python, and I don't > > > want to start writing the bindings for EWL from scratch without any > > > help. > > > > Did you ask for any? > > Did you propose any ? Why would I? Python is a shitty language, and I want nothing to do with it. But there is no logical validity to responding to my question with your question. It's nothing but a red herring. > We talked many times about this over #edevelop, and noone suggested > or said they could help. IRC is not always the best place to have that discussion. That's what mailing lists are for. If you didn't actively pursue help, you have no one but yourself to blame for not finding any. Well, and your poor choice of language. :-) > The only people interested in those bindings are doing them for ETK, > so that "help" I'm refering to will not come from cmarcelo or > barbieri, which is already an issue. It's best to concentrate all > our efforts in doing one thing. If you say so. > > > Can we please finally have an official, objective answer on this > > > very important matter, without partiality and without people > > > trolling one toolkit with false arguments only for the sake of > > > convincing us to choose their own toolkit. > > > > Nope. > > Thanks for trying your best to help the users. You seem to be under the mistaken impression that (1) the users care, which they don't; and (2) that declaring a winner will be best for users, which it won't. The people who care are the developers, and what's best for them in the presence of 2 similar libraries is that they choose for themselves. You are at a disadvantage because you're new. If you go back and read the archives, you'll discover that numerous warnings were given about the problems which would arise from having 2 widget libraries. raster opted for the current situation in spite of said warnings. I suggest you read these discussions before you continue with this thread so that less rehashing of past issues will occur. Michael -- Michael Jennings (a.k.a. KainX) http://www.kainx.org/ <[EMAIL PROTECTED]> Linux Server/Cluster Admin, LBL.gov Author, Eterm (www.eterm.org) ----------------------------------------------------------------------- "Or are you one of those folks who measures 'better' in terms of total installed base? In which case, the cockroaches would like to know when you'll be vacating their planet." -- Michael Paquette ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ enlightenment-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
