User talk:Quinyu

Several edits from historical sources
I believe several of the edits from historical sources should be referenced to the book source by cf. Also I believe some of these things are too detailed to include, giving history undue weight, as in article coniine. You added several formula containing e.g. B dot PtCl4, what is the B in this case? Is this some historical notation? Please check if this is to be converted to something more contemporary. The crystallizations of several alkaloids to characteristic double salts with Gold chloride or Platinum chloride are only of historical interest. Please don't forget the book source. I doubt if these additions do not go too far. 70.137.139.237 (talk) 20:53, 22 September 2009 (UTC)

Please reconsider undue weight of historical methods. Can you provide examples of outdated methods in the computer science section? As far as I know computer science has been pretty stable on its mathematical foundations for 30 years now. So I am interested in a discussion. 70.137.169.140

Outdated methods. Well, consider file systems.


 * Well, file systems still follow the general considerations as presented in the literature of the 70s.

Or for that matter, consider the entire x86 shebang, if you will: perhaps you know not of it, but the current processors are actually not natively x86 anymore. The processor needs to translate every single command into its own inner machine code first, and then it would execute that.


 * This is how microprogramming works since its invention in the 50s. I am very well aware of this and it has at least also some advantages for it. One is that you can still build the microcontrol store faster and wider on chip than off chip, and that the code memory bandwidth off chip is reduced by the more compressed format of the complex instructions vs. the expanded microinstructions.The other one is a stable user-level architecture which preserves investment into program development and user training.

Which, as a matter of fact, is overhead; it shows a few things. One, large developer companies are forcing the level of computer programming to stay, they are not yet ready to switch over to the new methods of programming that would be required to actually use the possibilities provided by the processor to 100%.


 * See above, it is not only overhead but a difficult trade off.

Two, the grand majority of programs is not yet multicore- and multiproc-ready, even though the technique had been available for ages; yet they do not bother to switch over to such way of thinking (and this is why, for example, the person who thinks he is a proficient programmer in VB or some other desktop-bound language, has to effectively relearn programming once he gets a job which requires him to write programs to run on a computer with 500+ CPUs) and that is why multitasking solutions are still inefficient; IPv6 has been available for ages, yet even most providers reject to switch over, let alone users. Why that? People stick with using inefficient general-purpose routines in GUI, in file system handling, in so-called "security" (which is not secure as much as they would like to think, in the end), instead of taking time to sit down and write their own routine that is polished to work close to the possible fastest and most memory-efficient way in one given situation, or do their homework and make a wider search about the most efficient way of solving that particular task.


 * You are missing some points here, it is again a difficult trade off and not as clean cut as you present it. The inefficient routines have the advantage that they are sufficiently debugged and offer a standardized interface. The first point gives a pretty good confidence in their functionality (does what it says on the box), the second one allows free exchange of modules using them, resulting in quick and dirty constructions and rapid prototyping from components out of the box.
 * The 500+ CPU systems are a different issue, many problems do not allow a decomposition into multiprocessor approaches, there is a whole grey scale for others problems between single processor solutions and massively parallel solutions, it all depends on the mathematical problem. Ordinary C-code usually allows rearrangement to a parallelism of 3 to 8 instructions in parallel in a mechanical fashion. Everything else requires for massive parallelism a deep analysis of the underlying mathematical problem and mathematical approach and is not successful on every problem. The single-thread one-processor solution is the most easily understood and the most generic one. If you get the solution to run with that in the prototyping phase, you can still later trade off development effort vs. targeted performance and scalability to multiprocessors if the need for performance arises at all. I am aware of the programming methods for super computers.
 * In security you have a similar trade off. Security routines like encryption methods and established protocols (AES, RSA, Diffie-Hellman, SHAxx) undergo a long public review in the security community and continued efforts by the hacker community and cryptographers to break them or to point at vulnerabilities.
 * To bypass this and roll your own will almost certainly lead to snake oil methods. If you want to roll your own, submit it to the security community for review and let it sit for 10 years, like has been done with RSA, Rijndael, Diffie-Hellman, DES/3DES, SHA.
 * Outside this scope you will absolutely be unsuccessful, trust me, I know what I am talking.


 * In the remaining points, motivation to use the old ones etc. I tend to agree with you. But I still have to remind you, not to give undue weight (compared to the bare bones essentials) to the historical methods. I am aware that not everybody has a GC/MS in the basement, neither do I.

Please contemplate that. However, why shouldn't you register an username while you're at it? it could simplify a number of things, including but not limited to user talk.


 * I only rarely edit in WP, and it works, as you see.

70.137.169.140 (talk) 11:55, 23 September 2009 (UTC)

No you are not misunderstood. I do not think that we can match the Roempp in detail. In the WP less is sometimes more. This is again a difficult trade off and not clean cut. This is what I meant by "undue" weight to historical procedures. But I certainly don't want to spoil the fun. But I also discussed your analogy from computer science, to hold, that this is not the question new vs. old and outdated, but a difficult trade off. Many things in computer science are not as outdated and historical as they appear. Insofar it is not a true analogy. 70.137.169.140 (talk) 12:15, 23 September 2009 (UTC) 70.137.169.140 (talk) 17:31, 23 September 2009 (UTC)

70.137.169.140 (talk) 12:15, 23 September 2009 (UTC)

Proposed Image Deletion
A deletion discussion has just been created at Category talk:Unclassified Chemical Structures, which may involve one or more orphaned chemical structures, that has you user name in the upload history. Please feel free to add your comments.  Ron h jones (Talk) 23:03, 10 June 2011 (UTC)

Notification of automated file description generation
Your upload of File:Abg-conhydrine.svg or contribution to its description is noted, and thanks (even if belatedly) for your contribution. In order to help make better use of the media, an attempt has been made by an automated process to identify and add certain information to the media's description page.

This notification is placed on your talk page because a bot has identified you either as the uploader of the file, or as a contributor to its metadata. It would be appreciated if you could carefully review the information the bot added. To opt out of these notifications, please follow the instructions here. Thanks! Message delivered by Theo's Little Bot (opt-out) 11:24, 7 December 2013 (UTC)

October 2014
Hello, I'm BracketBot. I have automatically detected that [//en.wikipedia.org/w/index.php?diff=630026117 your edit] to Omeprazole may have broken the syntax by modifying 1 ""s and 1 "[]"s likely mistaking one for another. If you have, don't worry: just [ edit the page] again to fix it. If I misunderstood what happened, or if you have any questions, you can leave a message on [//en.wikipedia.org/w/index.php?action=edit&preload=User:A930913/BBpreload&editintro=User:A930913/BBeditintro&minor=&title=User_talk:A930913&preloadtitle=BracketBot%20–%20&section=new my operator's talk page]. It's OK to remove this message. Also, to stop receiving these messages, follow these opt-out instructions. Thanks, BracketBot (talk) 20:50, 17 October 2014 (UTC)