[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]
[Off topic: lessons learned] Re: [xml-dev] Why is there an "S" inXSLT?
- From: Jonas Mellin <jonas.mellin@his.se>
- To: "Costello, Roger L." <costello@mitre.org>
- Date: Tue, 20 May 2008 09:48:24 +0200
Costello, Roger L. wrote, On 2008-05-19 17:55:
>
> David Carlisle wrote:
>
>
>> XSLT has two alternative names for the top level element
>>
[...]
> Write this:
>
>
> <?xml version="1.0" encoding="utf-8"?>
> <trn:transform xmlns:trn="http://www.w3.org/1999/XSL/Transform"
> version="2.0">
> ...
> </trn:transform>
>
>
> Comments?
I am relatively new to XML and while encountering XSL(T+FO) I believe I
experienced a similar feeling as you: why on earth have they called it
this? However, a few lessons learned from other areas kept me from
trying to change the world:
1) In active database research there was a discussion in late 1980 and
beginning of 1990 to call it reactive databases instead. The concept
reactive database would fit the idea better than the concept of active
database, BUT the concept of active databases had been around for
approximately 8-10 years with numerous research reports, funding
applications, numerous researchers etc. The decision was to stick with
the concept of active databases.
2) In real-time databases, we have been discussing using something else
than ACID properties of transactions, but we usually use ACID as a basis
and then tell the differences. The reason is that in databases, ACID
properties is a well know idea. Changing it to, for example, ACIM where
M stand for mission time rather than durability does not significantly
improve the situation.
3) Prof. David L. Parnas has described an interesting issue in the
american navy. He tried to change an error message on the screen in the
cock pit of their new fighters that were in the requirements engineering
phase while he was involved in the project. He was not allowed to. The
background story according to Prof. Parnas is as follows (if I remember
correcly):
The fighter air craft are fitted with two altitude sensors: one for
atmospheric pressure which is useful when you are flying upside down and
a radar which is useful when you are flying in the mountains. Since
fighters can be damaged, these two sensors can be broken. Currently, the
display then shows 3000 ft (or something similar) if they are broken.The
manual and instructions for flying the fighter then states: If the
displays shows 3000 feet ascend and if it still shows 3000 feet ascend
further. The figure 3000 feet is the average altitude that these fighter
fly on: a height which they rarely fly on. it came into being when a
programmer realized that the requirements in an earlier version of their
fighter realized that there was no requirement for the situation when
the height sensors were broken. The programmer did the right thing: he
phoned a user, a pilot in the navy. Unfortunately, he asked the wrong
question: "What is the average altitud of that the fighters fly on?".
The pilot, being a true military, did not question the question, but
answered after a few moments of calculations: "3000 feet". Thus, the
error message 3000 feet came into being. After this it has entered
manuals, classes etc.
4) A friend of mine works in a factory where they changed the
maintenance management system. The system cost approximately SEK
15'000'000. The initial costs for educating of the staff is
approximataly SEK 5'000'000 ( a few hundred employees went on training
sessions for a couple of days). He believed that it would take 1-2 years
before they would be using the new system properly.
The bottom line is: the world is full of idiosyncracies (as someone else
pointed out) for various reasons. The typical reason is that we cannot
foretell the future and when we name things it is difficult, if not
impossible, to put really good names on things. Sometimes we are lucky.
If a concept has been around for a while and been accepted by a large
user community, then it is difficult to change since the concept is part
of the legacy of the area. Changing such concepts cost money. In tthe
case of XSLT, there are numerious books, articles, web pages using the
prefix "xsl".
---
Carpe Diem!
===
Jonas Mellin, Assistant Professor in Computer Science
School of Humanities and Informatics, Building E-2
University of Skövde, P.O. Box 408, SE-541 28 Skövde, Sweden
Phone: +46 500 448321, Fax: +46 500 448399
Email: jonas.mellin@his.se, URL: http://www.his.se/melj
[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]