qp

My web development blog

Where I sometimes also write about other stuffs.


Messages my foot!

Since this website came online last year I have had a different quote/message in its footer every week. A little something to add some interest to the site. I sometimes go through the list because, clearly, I chose the quotes so I have some liking for their messages.

I just added a page listing all the footer messages used so far.
Permalink | Posted 3:19PM 6-4-2012 by Quentin.

Web dev matters -- "home menu" app

I am planning to create a simple web application to serve as a sort of homepage menu some time within this week or next.

This should be a fairly simple project, but just to make it more interesting: if I do not launch the basic app by midnight two weeks from now (15 April), I will eat my own cock.
Permalink | Posted 11:38PM 1-4-2012 by Quentin.

Site launch: Somewhy..?

"Somewhy" really needs to be in the dictionary.

Added a brief entry in portfolio for the site, which also happens to be the first entry! The first of many to come? Hopefully.
Permalink | Posted 3:56PM 21-3-2012 by Quentin.

When writing macros in Excel is good enough

Today's topic is about the use of Excel macros in contrast to fuller solutions that an IT guy or manager might propose for a given problem.

Mainly, Excel's VBA support gives the developer access to all the functionality of VB (the language, not attached to .NET, COM, or the like) along with all the basic functions available to an end-user of Excel. That is, a developer can programmatically sort Excel selections, create charts, etc, as well as make use of loops, conditional statements, file I/O and other programming constructs. Given these capabilities, Excel macros can accomplish many tasks for which one might be tempted to provide other solutions.

Perhaps then, given the breadth of Excel's VBA abilities, a more appropriate question might be when they should not be used.

When not dealing with lists of data
Where all the inputs and outputs involved are not lists of data, Excel should not be used. Running Excel puts an overhead over any programmed code and when a solution does not make use of Excel's spreadsheet functionality, it is simply unnecessary to have to deal with this overhead.

Where dealing with large/complex lists of data
Excel is fairly competent in working with lists of data -- being a spreadsheet program and all -- but macros written for Excel can still be slow. A developer might consider using a full database-backed solution where the data size is large or operations on the data are complex. One can imagine performing joins and unions on Excel data but such operations would certainly be more cumbersome to do in Excel when compared to the writing of SQL statements.

Many concurrent users
Any Excel solution will be limited by the fact that it is file-based and would have issues with multiple users accessing it at the same time. Therefore, any system to be used by more than about four people (personal opinion of course, choose a number you're comfortable with) or by people not within close proximity of one another should not make use of Excel. To elaborate on the close proximity point, the few users using a system need to be working close enough to do quick checks to ensure that others are not using it at the same time.

Where speed is critical
As previously mentioned, Excel places a runtime overhead on programmed code so any solution where speed is critical should not use Excel. People have a tendency to want speed in every technical solution they get, so the developer should make use of their discretion to discern the requirement for each case.

With the above considerations, an Excel macro-based solution would be appropriate for situations where simple manipulation of lists of data are involved, there are not going to be many concurrent users, and speed is non-critical. This still covers a wide range of scenarios, so developers might want to take Excel into consideration when they're thinking of options on how to solve a given problem.
Permalink | Posted 12:24AM 19-3-2012 by Quentin.

Sundry matters -- an operating system?

I am again toying with the idea of creating an operating system and trying to think up what this would require.

The obvious choices to me seem to be either to create the OS from scratch or to write a layer on top of an existing Linux distro. Given that I have zero experience with both, I suppose what I could do is to try both for a bit and see how that works out.
Permalink | Posted 3:17PM 4-3-2012 by Quentin.

displaying page 3 of 6
<<First <Previous | Next> Last>>