OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

SV: [xml-dev] Caught napping!



Hello again

I look at this thread with a bit of amusement. 

If reminds me a lot of one of the old Mac vs. PC/Windows threads, you
could pick up around 1994/5 on Fidonet. (And even earlier.)

I would like to come with some comments.

1. I agree that XML is great for documents, i.e. for handling the
exchange of a customer data as well as all books bought by that
customer. This is easiser than the CSV approach (one file for the
customer, one for the order heads, one for the order items, one for the
book items that are linked to from order items. (You dont have to make
that many levels, just overdoing it for the example.))

2. The item "Native XML Database" is not very well defined. Back when I
worked at Software AG, some prospects asked me: "So you actually store
all the XML documents as textfiles on the harddisk?" when I said "Native
XML".

3. We tend to forget that there is a whole lot of data out there,
allready stored in RDBMS (and IMS, somebody said). Those guys are very
happy about using XML for wrapping data for transmission, but they will
not give up their perfectly good RDBMS. (And hey, they can:
a. Generate the XML doc. on the fly (and even store the generated
XML-doc. in a log for later retrieval?)
b. Map received XML doc.s with no prob on the fly to RDBMS (and even
store the received XML-doc. in a log for later retrieval?)

4. You'all know Guy Kawasaki's "Top 10 lies of Entrepeneurs?"
http://www.garage.com/guy/speeches/Lies_of_Entrepreneurs.pdf ?

5. If I should envision a killer product in the XML storage arena, it
would not be a new "Native XML store". Why? Read "Crossing the chasm.".
"Native XML Database" is a disruptive technology, forcing the very big
majority to think in new ways. They will not like it. Instead a "XML
Frontend for RDBMS", that would XML enable RDBMS in a fast, easy and
convienient manner, and make it easy to store and retrieve XML in a
RDBMS, maybe using XML Schema for defining storage, and W3C XQuery for
retrieving data.

6. Remember 5-10 years ago, when you had to pay quite a bundle for a
TCP/IP network stack to your computer? 

7. I wouldn't have used this "suckerpunch", but now I am: Let us work
for decommoditizing the Internet, and bring back the excitement of the
"Netscape 1.1 beta" days.

Xtremely Mutual Love

Jens Jakob