jwcolby
jwcolby at colbyconsulting.com
Thu Nov 17 13:12:07 CST 2011
Your link takes me to a n email login screen in Russia. John W. Colby Colby Consulting Reality is what refuses to go away when you do not believe in it On 11/17/2011 2:00 PM, Salakhetdinov Shamil wrote: > Hi John -- > > Thank you for your clarification - from you original post > > http://e.mail.ru/cgi-bin/msglist?542573152=&folder=0#readmsg?id=13211081120000000542&folder=3 > > it wasn't clear for me that you're talking about particular client considering probable moving out of using MS Windows and MS database (Access or SQL) backends, and so I mistakenly implied you're going to rework you mighty "data crunching" system to use mySQL... > > -- Shamil > > 17 ноября 2011, 21:01 от jwcolby<jwcolby at colbyconsulting.com>: >> Shamil, >> >> If you read my original post closely you will see that it a client bringing up a new server, *not* >> my server here at my home office which is where I run the custom written C# program and SQL Server BE. >> >> As I think I stated in my original post, this client is at a transition point. They have to select >> an OS and a database engine. Their system uses Access backends, which we have split into many files >> due to size constraints. >> >> So for *this client* it is appropriate to study which route makes more sense. >> >> John W. Colby >> Colby Consulting >> >> Reality is what refuses to go away >> when you do not believe in it >> >> On 11/17/2011 9:34 AM, Salakhetdinov Shamil wrote: >>> Hi Hans, Jim, John, Darryl at all -- >>> >>> Sorry I didn't reply promptly on your postings here - I have had to release a solution to my customer... >>> >>> I do not state that mySQL or Linux are not worth to consider - that would sound really silly. >>> I'm just trying to keep to discuss this thread practical case when JC's is going to "jump at" mySQL after three years(?) of building his mighty computing environment with MS SQL being the core of it AFAIU. >>> >>> How much time it will take for JC to setup the new environment? >>> How much time the transition of JC's large database will take? >>> Will existing code work "automagically" with the new backend? >>> Aren't there other ways to use MS SQL on multi-core servers having just one core used for MS SQL and others for custom (code) data processing? >>> What "hidden side-effects" JC is going to meet during transition? >>> Who will pay (the bills) for that transition? >>> .... etc ... >>> >>> IOW there is no any "red herring" coming from here IMO. >>> >>> You know I'm not anymore that excited about learning new and new technologies as hobby - I'm much more excited about "getting things done" and my bills paid in time. >>> >>> I agree with Jim that these are interesting times our (custom) software development industry is entering nowadays - maybe even more interesting and exciting than 80-ies and 90-ies - but I'd disagree that without knowing Linux and mySQL and being "pure MS shop" one takes themselves at risk of getting out of market (let's not discuss that my point of view here as it could become OT with all the parties keeping their point of views unchanged....) - after all MS development tools cover/support most of the modern widely used desktop and web development technologies and web/cloud hosting costs are comparable for Windows and Linux and there are a lot of open source software and information for "MS-tied" developers... >>> >>> Hans you noted: >>> >>> <<<<<<<<<<< >>> But, in my opinion, more often than not, there are 3 essential reasons why people use the product they have: >>> >>> 1. They really need a certain feature >>> >>> 2. They are required to use that product (by a third party) >>> >>> 3. Or simply that they feel most comfortable using said product. >>>>>>>>>>>>> >>> >>> I agree but for JC real life case all the three above positions seems to be questionable IMO. >>> >>> As Darryl writes: >>> >>> <<<< >>> In reality we use a whole stack of software tools, From Excel, Access, SQL Server Express to dedicated mining software with GemcomWhittle and Surpac. The plan was not so much to replace SQL Server Express, but replace it for this job and see how it works. If MySQL has the grunt and space to get a result then I will use it this time. It may not be the best tool for next time. We really do use the best tool for the dataset we have. Sometimes that is Excel, although it usually is Access >>>>>>> >>> I agree of course but again for JC certain case IMO "hurrying to jump at mySQL backend" without trying to evaluate other alternatives while "staying at MS field" sounds questionable here. >>> >>> I can be wrong, life and JC practice will (soon) show was that worth to abandon MS SQL for ***one of his projects*** of not. >>> >>> Sorry John I'm referring your case - feel free to stop this discussion thread you started if you don't want to discuss it in this direction... >>> I could have misunderstood your intentions and very probably you're just going to "test the waters" there with mySQL and if it will not be "deep" for you then you'll make the full well planned transition paid by your customer - I'd have done the same... but after I'd have found I can't keep using MS SQL for a certain project/customer as it becomes too expensive to be true... >>> >>> Thank you. >>> >>> -- Shamil >>> >>> P.S. FYI: "Where Linux crushes Windows like a bug: Supercomputers" >>> http://www.zdnet.com/blog/open-source/where-linux-crushes-windows-like-a-bug-supercomputers/9890 >>> >>> 16 ноября 2011, 21:45 от Hans-Christian Andersen<hans.andersen at phulse.com>: >>>> >>>> Hi Shamil, >>>> >>>>> But "red herring" for me would be exactly stating that mySQL is a better alternative for MS SQL because it's "free" and open source, wouldn't it? >>>> >>>> I am not the sort of person who advocates that product A is better then product B, simply because it is free and open source. Likewise, I do not believe that product B is better than product A, simply because you paid for it. I believe that open source and proprietary software can co-exist. You just need to know what your needs are and what the pros/cons are to using either product. >>>> >>>> In my opinion, the point of open source is to raise the bar, so that proprietary software vendors feel the pressure to maintain their competitiveness rather than stifle markets with their monopoly. >>>> >>>>> I could be not quite accurate when talking about TCO, which for you doesn't include "hidden costs"(?) - for me that "TCO" does also include hidden costs. >>>> >>>> In that case, I will include the hidden costs. But I should warn you, it will get ugly! :) >>>> >>>>> Hans, I suppose, it's just partially true these days (as it's true for other companies including the ones, which develop and distribute open sources software) - they (MS) lock yourself in their technology as much as you wanted to be locked :) >>>> >>>> You will need to give me a good example of where this is true for open source databases. Both PostgreSQL and MySQL export to pure SQL dumped into a text file, rather than in a proprietary binary format. >>>> >>>> Btw, T-SQL is proprietary to Microsoft/Sybase, so no other product will be able to fulfill your needs, if T-SQL is what you absolutely must have. >>>> >>>>> I can tell the same from my and my customers experience: MS SQL and .NET Framework (all the main technologies within it) are very stable products, I have never had any problems that required urgent support if that only weren't a few bugs I occasionally left in my business applications solutions… >>>> >>>> If that's the case, then why pay for support? It would bring your TCO down and lower costs for your clients. >>>> >>>> If you want to seriously go over the TCO of MySQL vs MS-SQL or PHP/Python/Ruby/Whatnot vs C#/.NET, we could surely come up with some sort of conclusion. But, I personally don't think it would prove anything and it would be complicated (different TCOs for different use cases). When it boils down to it, they are all beneficial for their own reasons and, if they weren't, no one would be using them. >>>> >>>> But, in my opinion, more often than not, there are 3 essential reasons why people use the product they have: >>>> >>>> 1. They really need a certain feature >>>> >>>> 2. They are required to use that product (by a third party) >>>> >>>> 3. Or simply that they feel most comfortable using said product. >>>> >>>> - Hans >>>> >>>> On 2011-11-16, at 12:51 AM, Salakhetdinov Shamil wrote: >>>> >>>>> Hi Hans -- >>>>> >>>>> Thank you for your reply. >>>>> >>>>> But "red herring" for me would be exactly stating that mySQL is a better alternative for MS SQL because it's "free" and open source, wouldn't it? >>>>> >>>>> I could be not quite accurate when talking about TCO, which for you doesn't include "hidden costs"(?) - for me that "TCO" does also include hidden costs. >>>>> >>>>> And you approve my point of view/wild guess: "but if you start doing anything very particular, costs may go up and it may suddenly make more sense to go with Microsoft (it's always best to anticipate your needs in advance though!)." - don't you? >>>>> >>>>> "There are also a lot of negatives about going with Microsoft technology that you don't normally get with open source, for instance how I find Microsoft tends to try to lock you in to their technology" >>>>> Hans, I suppose, it's just partially true these days (as it's true for other companies including the ones, which develop and distribute open sources software) - they (MS) lock yourself in their technology as much as you wanted to be locked :) I mean that. Please do not accept me "locked by MS" - if I'd switch/use in parallel any other (development) technologies if that would be needed/requested by my customers and if that would be more profitable for my customers than using MS technologies.... >>>>> >>>>> "That said, MySQL (and other projects like PHP and so forth) are very very stable products. I've never had a problem that required support from MySQL." >>>>> I can tell the same from my and my customers experience: MS SQL and .NET Framework (all the main technologies within it) are very stable products, I have never had any problems that required urgent support if that only weren't a few bugs I occasionally left in my business applications solutions... >>>>> >>>>> Thank you. >>>>> >>>>> -- Shamil >>>>> >>>>> >>>>> 16 ноября 2011, 02:51 от Hans-Christian Andersen<hans.andersen at phulse.com>: >>>>>> >>>>>> >>>>>> Hi Shamil, >>>>>> >>>>>> I can't give you any exact figures for comparison, as I have never had to manage identical projects side by side to give a definitive account. Also, I would just like to point out that I am only really qualified to give you an answer that relates to web& mobile development. >>>>>> >>>>>> Personally, though, I think the topic of TCO is a bit of a red herring. Businesses typically buy products / packages, not databases and the nuts& bolts so to speak, and therefore they are purchased through vendors, who then provide the support for the product and each component in the system. That is, unless you happen to be a company that sells a product that relies on MySQL or whatnot, in which case, MySQL and such will typically have support licenses. Not only that, but many 3rd party companies exist to provide support offerings, so there is that competitive element to it. >>>>>> >>>>>> That said, MySQL (and other projects like PHP and so forth) are very very stable products. I've never had a problem that required support from MySQL. >>>>>> >>>>>> Finally, the cost of development is largely going to be about salaries. I've looked at sites like itjobswatch.co.uk and found that in almost every category, Microsoft developers / admins / etc demand a higher price. There are a few notable exceptions, but then you realize "oh, when they are talking about Linux, they are also talking about kernel developers, while for Windows, it is simply just Windows certified admins". So some things can't really be directly compared. >>>>>> >>>>>> Licenses, well, open source software is free, so you can guess how that compares. >>>>>> >>>>>> Ultimately though, I just want to say (and the point I was trying to make in my earlier email) is that this discussion really revolves around the type of software / product we are talking about. TCO for an open source project like MySQL may be extremely low compared to Microsoft in most cases, but if you start doing anything very particular, costs may go up and it may suddenly make more sense to go with Microsoft (it's always best to anticipate your needs in advance though!). If you want to talk about specific numbers, we probably also need to be specific about the scenario. >>>>>> >>>>>> There are also a lot of negatives about going with Microsoft technology that you don't normally get with open source, for instance how I find Microsoft tends to try to lock you in to their technology (ie. I found exporting data from MSSQL to MySQL to be quite a hurdle, while the importing tools from MySQL to MSSQL to be rather good, hah), but I'll leave that to another discussion as it doesn't directly relate to TCO (but it's a hidden cost, in my opinion). >>>>>> >>>>>> - Hans >>>>>> >>>>>> On 2011-11-15, at 9:58 AM, Salakhetdinov Shamil wrote: >>>>>> >>>>>>> Hans -- >>>>>>> >>>>>>> Than you for your remark. >>>>>>> >>>>>>> Sorry, but I'm talking about "Total Cost of Ownership" (TCO) (long run costs - license, development, support etc.) >>>>>>> >>>>>>> .NET/C# (VS2010) + MS SQL vs. mySQL + ... (?)..... >>>>>>> >>>>>>> I haven 't seen any real figures - did you? >>>>>>> >>>>>>> My feeling is that .NET/C# 4.0 (VS2010) + MS SQL TCO is (considerably) lower even with that high license costs as JC quotes... >>>>>>> >>>>>>> Thank you. >>>>>>> >>>>>>> -- Shamil >>>>>>> >>>>>>> 15 ноября 2011, 12:51 от Hans-Christian Andersen<hans.andersen at phulse.com>: >>>>>>>> >>>>>>>> >>>>>>>> Shamil, >>>>>>>> >>>>>>>> Of course you need to evaluate what product you are building to decide what the cost vs. benefits are. However, I would just like to add that Qt is an excellent development platform and, not only that, but it is also natively cross platform (without being java! ha!). But there are, of course, many benefits to using Microsofts "platform" of tools. The unfortunate truth is that some applications still need to be made native (or close to it) and Microsoft still has a stranglehold in the desktop market, so going in that direction can still make a lot of sense. That's the world we live in now though. No longer is there a simple solution to everything, but many different angles to approach a problem and you need to be well versed enough to know which is best to take and what the pros/cons are. >>>>>>>> >>>>>>>> - Hans >>>>>>>> >>>>>>>> On 2011-11-14, at 11:28 PM, Salakhetdinov Shamil wrote: >>>>>>>> >>>>>>>>> Jim, John at all, >>>>>>>>> >>>>>>>>> That's is an interesting discussion. >>>>>>>>> But did you ever try to compare development and support costs of >>>>>>>>> >>>>>>>>> MS SQL + .NET vs. mySQL + ...? >>>>>>>>> >>>>>>>>> I must note I know almost nothing about Linux development tools, but I have worked in the past for a long time on IBM360/370 and PDP11 RSX-11M so JCL, command line, and macro-assembler (especially great for IBM360/370) and PL./1 and and pure C or C++ or Pascal development with file systems or network (CODASYL) or relational databases without rich backend-level data definition and manipulation tool - I have my own experience to evaluate how much time all that development takes... >>>>>>>>> >>>>>>>>> .Now take >>>>>>>>> >>>>>>>>> C# 4.0 (VS2010) + MS SQL with T-SQL with SPs and UDF vs. mySQL + ... >>>>>>>>> >>>>>>>>> - wouldn't MS SQL license costs be covered manifold by higher Linux development and support costs? >>>>>>>>> >>>>>>>>> And I'm not talking/meaning Web site development where becoming standards HTML5 + CSS3 + JavaScript would make Windows and Linux development and support costs comparable(?) I'm talking about development of desktop custom business applications... >>>>>>>>> >>>>>>>>> Thank you. >>>>>>>>> >>>>>>>>> -- Shamil >>>>>>>>> >>>>>>>>> >>>>>>>>> 15 ноября 2011, 08:13 от "Jim Lawrence"<accessd at shaw.ca>: >>>>>>>>>> John: >>>>>>>>>> >>>>>>>>>> All I can say is talk to Hans as he has made it his career to work with >>>>>>>>>> Linux systems (about 10 years now) and I am sure he would help in the case >>>>>>>>>> that you ever became completely stuck. >>>>>>>>>> >>>>>>>>>> It is nice of me to volunteer his services like that but he will understand. >>>>>>>>>> ;-) >>>>>>>>>> >>>>>>>>>> To be honest, Linux is not as difficult as it first looks. My next client is >>>>>>>>>> asking for an inexpensive alternative to their current Windows system as the >>>>>>>>>> prices, for the same, have slowly crept beyond their ability to pay. I have >>>>>>>>>> suggested Linux and they are currently reviewing it and so far they have >>>>>>>>>> encounter no real reasons to stay put. (I am paid a flat rate so >>>>>>>>>> implementation will not be any more expensive than the current contract.) >>>>>>>>>> >>>>>>>>>> They have already, cautiously moved their MS Office to Open Office and have >>>>>>>>>> found they had little trouble adjusting. >>>>>>>>>> >>>>>>>>>> Jim >>>>>>>>>> >>>>>>>>>> ) >>>>>>>>> >>>>>>>>> -- >>>>>>>>> AccessD mailing list >>>>>>>>> AccessD at databaseadvisors.com >>>>>>>>> http://databaseadvisors.com/mailman/listinfo/accessd >>>>>>>>> Website: http://www.databaseadvisors.com >>>>>>>> >>>>>>>> -- >>>>>>>> AccessD mailing list >>>>>>>> AccessD at databaseadvisors.com >>>>>>>> http://databaseadvisors.com/mailman/listinfo/accessd >>>>>>>> Website: http://www.databaseadvisors.com >>>>>>>> >>>>>>> >>>>>>> -- >>>>>>> AccessD mailing list >>>>>>> AccessD at databaseadvisors.com >>>>>>> http://databaseadvisors.com/mailman/listinfo/accessd >>>>>>> Website: http://www.databaseadvisors.com >>>>>> >>>>>> -- >>>>>> AccessD mailing list >>>>>> AccessD at databaseadvisors.com >>>>>> http://databaseadvisors.com/mailman/listinfo/accessd >>>>>> Website: http://www.databaseadvisors.com >>>>>> >>>>> >>>>> -- >>>>> AccessD mailing list >>>>> AccessD at databaseadvisors.com >>>>> http://databaseadvisors.com/mailman/listinfo/accessd >>>>> Website: http://www.databaseadvisors.com >>>> >>>> -- >>>> AccessD mailing list >>>> AccessD at databaseadvisors.com >>>> http://databaseadvisors.com/mailman/listinfo/accessd >>>> Website: http://www.databaseadvisors.com >>>> >>>> >>> >> -- >> AccessD mailing list >> AccessD at databaseadvisors.com >> http://databaseadvisors.com/mailman/listinfo/accessd >> Website: http://www.databaseadvisors.com >> >> >