Re: New RDBMS implementation

On Wed, 05 May 2004 00:16:21 -0300,
=?iso-8859-1?q?Leandro_Guimar=E3es_Faria_Corsetti_Dutra?= <leandro_at_dutra.fastmail.fm> wrote:

>Em Tue, 04 May 2004 16:46:57 -0700, Alfredo Novoa escreveu:
>
>>> Not yet. It has been requested, and if I convince the bosses
>>> here we'll create one. No idea how soon yet.
>>
>> Do you have info about how to write a Dataphor device?
>
> Not specifically, but I remember checking some existing ones
>quite some time ago and it didn't seem specially difficult.

But have you the code of the existing ones?

>>> > You can build a complex script and to send it to the DBMS.
>>>
>>> Yep... what I do mean is when you want to do something like
>>> SQL embedded in COBOL or C.
>>
>> It is the same.
>
> Do you mean sending COBOL or C to the DBMS? You lost me
>here...

No I mean sending Tutorial D from C# to the DBMS.

query.Exec("update wages where name = 'Alfredo' set {wage = wage * 2}");

> Check their release notes for details, but they decided the
>type hierarchy wasn't worth the trouble. Looks like their informal,
>virtual review board made by Date, Darwen and others didn't protest,
>at least not loudly.

For typical business systems probably not. All we really need are chars numbers and booleans. But for a general purpose DBMS it would be very recomendable IMO.

>> It would be possible to create a Java/C coders friendly version of
>> Tutorial D and another for the Basic or Pascal coders. VB .Net and C# have
>> almost indentical semantics, to chose one of the other is a matter of
>> taste. We could do the same with Tutorial D and D4.
>
> That'd be nice. I wasn't so much thinking about Tutorial D
>and D4 interoperability, since Pascal isn't so much more respectable

There are still millions of Pascal hackers.

>than IBMese in hackerdom, but about something C-like and something
>functional perhaps.

But not strictly functional. To drop relvars would be a big issue :)

> Nathan thinks similarly to you here, perhaps even more
>radically: if I remember well, he said 'anything that generates the
>same compiler tree' or something to that effect.

It is the same I think. A parser is a very little and simple part of a DBMS. If two languages are semantically equivalent it is easy to generate the same intermediate representation tree with both. Then you would not need to modify a single line of existing code to add a new language.

>>> Dunno. Have you seen what Paul Graham did in what has became
>>> Yahoo! Stores?
>>
>> No, what he did?
>
> It's all Lisp. Also Sabre seems to be 100% Lisp.

But Lisp is not a relational database language.

>>> You are painting yourself in the MS, proprietary corner... you
>>> can't win at MS game.
>>
>> Visual Studio is a great IDE and I don't pretend to put MS out of the
>> business :)
>
> Yes, but they will put you out if you turn to be a competitor.

They can do that anyway. I don't see anything wrong in VS integration except that it is still very cumbersome. I readed about they are creating a new language infrastructure for dummies called Common Compiler Infrastructure.

> Haven't you heard the rumour that in the US venture
>capitalists won't fund competitors to MS, because the risk of being
>stamped out instead of bought out by them is too big?

It would be a suicide to compete with MS with a word processor, but they are nobody in business systems. Navision is very mediocre and has a primitive navigational technology.

> Or that at a
>MS-WCE conference there was a line to the questions microphone where
>people wanted to describe their projects and ask MS representatives if
>they didn't intend to enter that particular niche?

What I hear is that Microsoft made a mistake with Navision.

>> SharpDevelop is not a bad option. Eclipse could be another, but I don't
>> know it.
>
> I had forgotten about #Develope, but Eclipse is Java which you
>dislike.

SharpDevelop is still very far from VS, and it is true that it would be weird to use Eclipse in a Windows enviroment.

>>> So what? It just proves almost everybody you know is in the
>>> MS camp.
>>
>> And it would not be very intelligent to ignore that fact.
>
> Not ignoring, but ensuring you get portability and less
>competition. And freedom from patents, have you seen /. today?

But also a little fraction of the potential customers. Our customers are little and mid sized companies and there MS has practically the 100% of the market.

If Mono or dotGNU were able to run MS Office things could change.

> One takes advantage of source code not by changing the OS but
>by having the possibility of doing that.

But it has not practical value in almost all cases.

You can count with your fingers the people that touch the Linux kernel.

>> Java is still more popular but .Net framework is superior (has more
>> quality) in almost every aspect, and it is maturing a lot faster than Java
>> did.
>
> Quite obvious, being a later arriver enabled it to learn from
>Java.

To learn and to do the things a lot better. Many people describe the Net as a better Java.

> Java developers seem to disagree quite a lot from you.

I also developed in Java.

>Specifically integration with MS products can be counterproductive as
>it may mean strengthening proprietariness and a competitor.

But many customers demand integration with MS products.

>> Portability is not a big concern for us because Windows is everywhere in
>> our sector and the MS guys are porting the framework to other hardware
>> platforms. Light fast rich graphical user interfaces are the key.
>
> Hardware platforms are not significant, they exist only in the
>MS WCE space, and that's for PDAs, WinTerms and the like.

But this is the raison d'etre the .Net framework. The 64 bit "revolution" is near.

>> Well the problem is that I have not a clue about how they manage to make
>> money with open source. Where can I read about it?
>
> Hm, you can start with some articles about Zope or MySQL that
>recently made the rounds at /. and LWN, respectively
>http://slashdot.org./ and http://lwn.net./. The website for GNU Ada
>is in fact the one for the company maintaining and servicing it, and
>it seems to be quite popular with US gov contractors. Information
>about the GNU toolchain development and support has been harder to
>find since Cygnus merged with Red Hat, but they were known to be quite
>lucrative -- it is extensively used in the embedded market, and
>support contracts are quite pricey --, the same being true of
>ArsDigita in the collaborative website development space. I thing the
>website for Ghostscript is yet maintained by Aladin Software or
>something the like (*not* the makers of compression utilities), in
>fact its author earned early retirement licensing and supporting it
>for Postscript-compatible devices.

Thanks, I will look this.

Regards
  Alfredo Received on Thu May 06 2004 - 13:16:47 CEST