1. Skip to navigation
  2. Skip to content
  3. Skip to sidebar
Source link: http://archive.mises.org/10792/mises-org-needs-database-administrator-help/

Mises.org needs database administrator help

October 7, 2009 by

The popularity of Mises.org has been skyrocketing over the last several years, and with it, so has the load on the software and hardware running this site. While we’ve held up the hardware and website part of it, we’ve never had a database professional to take a good look at the data behind Mises.org. If you’re a Sql Server expert and would like to help me rewrite some very tricky SQL queries at the heart of Mises.org, please contact me. One day, Jeff Tucker will write an epic article about all this – this is your chance to be part of the story.

{ 22 comments }

Giant_Joe October 8, 2009 at 12:17 am

What kind of SQL are we talking? I’ve done MySQL, some PostgreSQL, and some Oracle db stuff (a long time ago)

Jordan Bullock October 8, 2009 at 12:17 am

I consider myself more intermediate than expert, but I have experience with MSSQL if you’re looking for assistance. Feel free to email me.

David Veksler October 8, 2009 at 12:39 am

This is for Microsoft Sql Server. Take a look at the bottom of this page: http://mises.org/Community/blogs/misestech/archive/2007/10/09/tagging.aspx

jgabios October 8, 2009 at 1:55 am

As I understood from your posts, you guys here at Mises.org are against IP, and I like that. Instead of fixing some sql queries for MSSQL, i suggest you move to postgresql, or other free software db engine, you would find better support then. i guess no need to argue here where free market feels at home:in the free software universe or proprietary software universe.

David Veksler October 8, 2009 at 2:02 am

jgabios,

Putting aside arguments about whether open source or proprietary software is better, if someone is willing to donate their time to create new functionality for the Mises Institute, we would love to have them. So far, only people who use Microsoft technologies have created new functionality for Mises.org.

Maybe that is just a coincidence, or maybe that says something about (1) how productive you can be with proprietary software and (2) how some people focus on getting things done rather than debating what system is best.

newson October 8, 2009 at 4:28 am

…and so the glove is thrown down!

Jack October 8, 2009 at 6:22 am

I also have experience with MSSQL (not a supreme expert but few years experience), so feel free to contact me if you need help.

Shane October 8, 2009 at 7:25 am

I have quite a bit of Experience with SQL Server.. Let me know how I can help.

Andras October 8, 2009 at 7:28 am

While I am not a DB specialist (I am a programmer/developer) I do some T-SQL on and off… anyway here is the question:

‘The problem is that it currently ranks tags according to their total popularity, not their “related-ness.”‘

The question is, do you have some “mechanism” in place for tracking “related-ness” because I don’t think it is possible to solve this problem purely with SQL if you don’t have an algorithm to grade the level of “related-ness” of the given tag to other tags and insert/update that info to the database? In other words, somehow you should be able to create and maintain information about “related-ness” before you can use that information in your SQL queries.

Steve October 8, 2009 at 8:12 am

I also have considerable experience with MS SQL server and I’d love to help out the Organization.

Ron October 8, 2009 at 8:39 am

I’m a Senior SQL Server DBA, and I’d be glad to help out! Shoot me an email if interested.

Mike October 8, 2009 at 9:00 am

I haven’t donated to Mises but I’d love to donate some sweat equity (of sorts). I’ve been involved in the web programming and DBAing scene for some 10 years now, exp. with MySQL, Postgres, Oracle and MSSQL.

I would suggest that you setup some kind of a forum somewhere not necessarily here and post some code snippets and database schemas so that the half-dozen or so people who have expressed interest in helping can all give it a whirl.

Cory October 8, 2009 at 12:09 pm

I think Mike’s suggestion is a good one. I work for one of the largest domain services providers providing data analysis work. Our data warehousing servers are MS SQL. We have terabytes of data and writing SQL for transactional databases is much different than for standard relational data storage. However, I haven’t done traditional DBA work in about 2.5 years, so while I may be able to provide input on query efficiency, data transformation and loading, someone might be much better at discussing the ins and outs of tuning SQL Server 2008 for performance. This would give everyone with the needed expertise to not only review the problem, but also comment on the +/- of the potential solutions. Online code review!!

Nick October 8, 2009 at 5:13 pm

Andras and Mike’s comments together are pretty close to what I would say. Without knowing the schema, it’s hard to say if what you’re trying to do is currently possible. If [tagmap] only relates one tag to another and [tag] doesn’t have a field to group tags then it might not be possible without schema changes. However, if the [tag] table contains a field that can group tags (ie; [tag].[parent_tag_id]) then the query can be altered to order the related tags based on their parent tag (assuming tags that share the same parent are always more related than tags that don’t). Of course, this is all speculation. As Cuba Gooding said, “Show me the schema”.

Peter October 8, 2009 at 6:59 pm

Maybe that is just a coincidence, or maybe that says something about (1) how productive you can be with proprietary software and (2) how some people focus on getting things done rather than debating what system is best.

The answers to both of those are pretty clearly enormously in favour of free software. But how does someone not already inside Mises.org go about “creating new functionality for Mises.org”? And especially using non-MS products when you run MS software?

Andras October 8, 2009 at 7:46 pm

Peter:

I think that most of the future will be open source software but I also think proprietary will be present because in freedom one is free to have secrets too (it is another thing if one use government granted monopoly which is sometimes the case) and because it is virtually impossible to have open source without a thin layer of proprietary under. Just like it is impossible to have industry without agriculture and knowledge production without industry. The proprietary model is the industrial model of software development and delivery, it won’t die it will just became smaller and smaller part of the whole software business.

Anyway proprietary software is not a bad thing per see, it is a productive contribution to economic exchange. The restrictions imposed by the vendor is acceptable most of the time because the value provided is much better. It maybe doesn’t help you as open source does but it doesn’t hurt you too (except if they try to enforce the government granted monopoly “rights”… the practice of this was at least until now mostly nonexistent). I am against software patents, but proprietary software is possible without software patents too and most of the proprietary software doesn’t hold patents, they just choose to keep the source code secret. A free choice people should have.

The fact is that the only real alternative (in terms of features and usefulness) to MSSQL is Oracle and IMBDB2 and both are proprietary, and compared to any of those MSSQL actually looks like free :)

Does Mises.org need those features? that is up to the people who design this application, but I am sure they don’t have to go with the less capable product just because it happens to be open source. Another thing is that open source is actually not free (as in beer) the economic exchange in the open source model is via code not via money, and while everybody can use it and nobody have to pay for it, the developers doesn’t write open source software because they want to give free software to the market, they write open source software because they want to share their work with other developers… but this is a long story… the point is that open source is one of the first business models of the knowledge economy (the economy where most of the growth will come or is coming from knowledge production and from production of other non-rival economic resources) that is why it is very easy to see it at a wrong way, and very hard to understand that model and its implications. That is why the IP based business model which worked for the industrial age (I don’t say it was good but it worked) is so counterproductive for our age and will be absolutely impossible 10 or 20 years from now.

Kevin Hodgkins October 9, 2009 at 2:53 pm

Jeff,

I would be more than happy to help out. I live in Birmingham so could even come down if needed. I have 10 years of SQL and BI experience with MS SQL. You may want to think about Analysis Services queries within SQL that has some great “real-time” data mining capabilities that are good at suggesting relatedness (kind of like what Amazon does to relate possible items for you).

Peter October 10, 2009 at 5:55 pm

it is virtually impossible to have open source without a thin layer of proprietary under.

What makes you think that?? I don’t have a problem with proprietary software; but you really want to have open source underneath! And there’s really very few things (for most people: nothing at all) that isn’t done as well or better with free software nowadays.

The fact is that the only real alternative (in terms of features and usefulness) to MSSQL is Oracle and IMBDB2 and both are proprietary, and compared to any of those MSSQL actually looks like free :)

MySQL is a toy, but PostgreSQL isn’t.

Andras October 10, 2009 at 7:32 pm

Peter:

I like open source and I use it and I even contribute to it… but I don’t think proprietary is evil. Anyway to answer your question: “underneath” I didn’t mean that some actual open source software need a proprietary software layer under (sorry for my expression), I mean that the open source software “industry” need a parallel proprietary software “industry” because without it it is impossible to have open source at all. There are many kind of software codes, some of it is interesting to develop (hot) some of it isn’t, millions of lines of code simply wouldn’t be developed or maintained if there is no proprietary model with guaranteed profit in monetary terms. I am sure most of the popular (hot), software is available in the open source model, but not all the software are popular and hot nor the problems we have to solve are all interesting so that the open source community would provide a solution for it. There are some jobs that must be done and people will do it only for the money. Open source requires commitment, passion and if you want even a form of love, many problems cant attract developers because it is impossible to have passion and all the required stuff toward them.

And last and not least, only a fraction of the software we have in this world runs on personal computers or general purpose servers, most software runs as embedded applications in millions of devices, and while some of the libraries are open source not everything is possible to develop by that model.

MySQL is not a toy but not so powerful and such feature powerhouse as MSSQL, Oracle or IBMDB2… PostgreSQL is in some areas and for some purposes better than MySQL and in some areas and for some purposes not, so generally it is the same league as MySQL.

David V October 14, 2009 at 12:54 am

Hi to everyone who volunteered:

I have made a copy of the primary Mises database, scrubbed it for confidential data and put it here along with a list of priorities:

http://mises.org/Community/wikis/ideas/database-information.aspx

There is lots to do, and I will find more if you get through the current list.

Please contact me directly if you need more permissions.

The Mises.org C# codebase is open source. Access information: http://mises.org/Community/wikis/ideas/mises-org-access-information.aspx

David V October 14, 2009 at 1:21 am

“But how does someone not already inside Mises.org go about “creating new functionality for Mises.org”? And especially using non-MS products when you run MS software?”

Mises.org is open-source and in addition to .Net runs MySql, Perl and PHP. There is a long list of features we’d like, so as long as your language of choice can work with MS SQL, you can work on a feature (assuming it is a standalone page and not a change to existing .Net pages) and send the code to us.

Peter October 18, 2009 at 1:18 am

Anyway to answer your question: “underneath” I didn’t mean that some actual open source software need a proprietary software layer under (sorry for my expression)

I interpreted it as meaning you wanted proprietary software for the OS and system libraries, underneath whatever open software you run on top; which is precisely the opposite of what I would say!

I mean that the open source software “industry” need a parallel proprietary software “industry” because without it it is impossible to have open source at all. There are many kind of software codes, some of it is interesting to develop (hot) some of it isn’t, millions of lines of code simply wouldn’t be developed or maintained if there is no proprietary model with guaranteed profit in monetary terms.

Oh; you’re confusing open source with freely-developed — not the same thing at all. There’s no reason open source software can’t be written for pay (in fact, most of it is!)

There are some jobs that must be done and people will do it only for the money.

And they’ll pay for the tools they need developed to do that job…but it doesn’t mean the tool has to be proprietary!

And last and not least, only a fraction of the software we have in this world runs on personal computers or general purpose servers, most software runs as embedded applications in millions of devices, and while some of the libraries are open source not everything is possible to develop by that model.

Sure; but software on those devices is even more valuable as open-source, and there’s even less reason for the producers to want to keep it closed-source, since they control the hardware (though precious few of them realize that); but that’s another subject.

Comments on this entry are closed.

Previous post:

Next post: