May 28, 2003
Data Modelling Tools
I'm in the market for a cheap (well alright, free) data modelling tool. I want one that allows me to draw pretty pictures and from the generate valid DDL that can be used to build a schema. It would be nice if if could reverse engineer existing schemas.
For added brownie points I should be able to specify a logical data model (entities, attributes and relations). For extra, extra brownie points it should allow me to link logical and physical data models.
My databases of choice are Oracle, DB2, SQL Server and MySQL. I don't mind if it doesn't support all of these out of the box, as long as I can amend the code generator for the particular quirks of each environment. Oh, and did I mention that it should be free?
I had a look at Google, but it wasn't much help.
I'd rather not write one myself, as that could take decades, but if you know of a project that is close to my needs then fire up the comments below.
May 16, 2003
Why Use One Word?
Spotted in an email from my project manager this morning;
"With respect to the Web Workshop on Tuesday, we need to decide how to support the process from a Technical perspective. Arguably <developer> should be present, but that will impact the <hand waving> stream... we'll catch up on this one."
The scary thing is that he talks like that as well. That project motto is so very apt.
May 13, 2003
We have recently adopted a new one;
"I'm a developer, get me out of here."
For the overseas visitor, perhaps this web site might provide a little illumination.
Data Access Objects
A quick Google doesn't find anything and a scant glance at the OR mapping toolkits I have found so far doesn't show anyone claiming to have implemented this particular pattern.
I'd rather not re-invent the wheel but I'm going to have use this in Java at work over the next couple of months. To work out my understanding I'm going to first write the code in Python, hence this request.
May 12, 2003
Last Blogging Tools Entry
For now anyway.
I guess I just have a slightly different usage model and find that I don't rely on directories nearly as much as I do the ability to search. Having a user defined classification scheme is a bit of a mixed bag for me as I find it easier to remember what I have posted rather than which category I've placed it in. I've observed that some people have more of an associative memory rather than accessing things through indexes.
Perhaps an example will make my point more clearly. Since I've been surfing the web (only about five years, what a lightweight) I've been collecting bookmarks. Every time I get a new machine (or a new browser) I find myself setting up a slightly different hierarchy of folders, usually with duplicates or at least some areas of overflow. This is a reflection of my changing perceptions rather than anything else. But it does confuse me.
For instance, where did I put that link to Andrew Wiseman's Television Room? Was it under 'Fun', 'Leisure' or 'Information'. Each of those has a 'TV' subfolder on my different machines and it could be in any. Actually, it wasn't stored on my current working machine as I haven't looked at that particular site for a while. So I did what any surfer would do, I went to google and entered '625' into the box. Bingo, 1st entry.
Which is a fair approximation of what happened when I experimented with categories on my blog. I ended up with one for technology, one for python and one for databases. Then I wrote an entry on MySQLdb. Which really could live in any or all of these. So I threw my hands up and don't have any categories. Which is why I suggested a search engine may be more useful to me.
My point, I guess is that the ultimate weblogging system should support both umpteen levels of classification and a search facility.
Still, the continuing discussion is fascinating and shows that Matthew has touched a nerve. I like his shopping list and only have one more thing to add. This probably is too detailed to really make the function outline, but I don't want to have to type and edit my posts in a web browser. Its a lousy writing environment and I'd much rather use other tools and produce a file for each entry which I could then upload to create a post.
Bring me an attach button or bring me death. Well, at least in the current interface anyway.
May 02, 2003
More blogging toolage
Of course, just after I wrote my last entry I came across mpt's shopping list for blogging functionality. All good stuff, and I especially like hist RESTful approach to URIs.
The only thing I am not mad keen on is his very complicated category and facet classification scheme. I appreciate that you can accumulate a lot of content over the years (even this blog has 114 entries). But rather than classify it in ever more obscure ways why not just add a search engine? A classification system is quite useful and should be part of any tool, but perhaps not as complex as the one proposed.
A new blogging tool
Somewhat overtaken by other events, Michal Sabren proprietor of Cornerhost has let slip about rantelope a new blogging tool. It is being written in Python (which is a good thing) and aims to combine the best content management and weblogging features of the market leading products.
He has mentioned his intention to open source the project and build a community around it, which is also a good thing. There is no publicly available code at the moment though.
What is not immediately apparent is what differentiates rantelope from the other tools out there, such as Movable Type, Blogger, Greymatter, Radio and Blosxom. Not to mention some of the less mainstream tools like pybloxsom, textbox. Still, I think that its worth keeping an eye on it.