Journal Articles

CVu Journal Vol 12, #6 - Dec 2000 + Letters to the Editor
Browse in : All > Journals > CVu > 126 (17)
All > Journal Columns > LettersEditor (132)
Any of these categories - All of these categories

Note: when you create a new publication type, the articles module will automatically use the templates user-display-[publicationtype].xt and user-summary-[publicationtype].xt. If those templates do not exist when you try to preview or display a new article, you'll get this warning :-) Please place your own templates in themes/yourtheme/modules/articles . The templates will get the extension .xt there.

Title: The Wall

Author: Administrator

Date: 08 December 2000 13:15:40 +00:00 or Fri, 08 December 2000 13:15:40 +00:00

Summary: 

Body: 

A Request, An Enequiry, A Suggestion

Dear Francis

1. For the past few months I have been frustrated because I don't know (and [don't want][have no time] to learn) C++, and can therefore not participate in the "student code critique". Since C++ is a superset of C, it would surely be more equitable to have a ratio of C to C++ competitions that reflected this? Alternatively, why not have one for each language every month? There is plenty of shonky code out there to choose from. For instance, you (and others) are always going on about how bad Herbert Shildt is - why not use some of his code (with suitable acknowledgements, of course)?

2. I have been labouring under a misapprehension ... or have I? I have recently been moved to a project which is upsizing to SQL Server (from Access) and since this will now be globally distributed we needed to use GUIDs (Globally Unique Identifiers) to ensure data integrity. My understanding is that these are constructed from a combination of timestamp (guaranteed no rollover until approximately 3,400 A.D.) together with the IEEE 802.1 node address (a 48-bit number), which I had thought was analogous to a serial number hard coded into the network card. If no network card is available, then a cryptographic quality 47-bit number will be obtained, with the MSB of the first octet of the node ID set to 1, which is guaranteed never to conflict with an IEEE 802 address. If this is correct, I find it hard to envisage a scenario where data integrity might be compromised due to duplicate primary keys, which seems to be one of the (many) thrusts of Silas Brown's article "Visual Control and COM". Any experts want to settle this? I have trawled the web but could find nothing that my puny brain and time resources can master.

3. In my last job (some four years ago, so the details are rather hazy) my employer had a publication division, and during my time there I had to typeset four books (all of 150+ pages with copious illustrations and tables) using Quark Xpress™ on a Mac. The copy was submitted in Microsoft Word™ PC format, and I had a devil of a job attempting to import it with the Word formatting intact. If you (the editor) are having similar trouble (as you have alluded), why not do what I did, save all your copy as ASCII text, import it as such, and then use Quark's own formatting entirely. Incidentally, one of the things I noticed about habitual Quark users was that they rarely used the menus, but preferred to use the many (some very complex) combinations of hot/meta keys, a habit that I wasted no time in acquiring myself. I now find it painful to watch my colleagues working in Word, for example, selecting (a) word(s) by dragging with the mouse (double-clicking on a word selects the whole word, treble-clicking selects the whole paragraph. Quark XPress™ is even better, as you can select word/sentence/paragraph/whole document by 2, 3, 4 or 5 clicks), locating the Edit menu, then clicking on Copy, or right-clicking the mouse button and scrolling down the list, when CTRL+C or ALT+E, C will do it in no time at all. But then I find it painful to watch people doing two-fingered typing while looking at the keyboard when tHIS SORT OF THING CAN EASILY HAPPEN. Aren't I smug? Seriously, though, the failure of professional programmers to master touch-typing, when the keyboard is arguably their primary tool, strikes me as completely absurd, and something I rant about fairly often. It takes only an hour or so a day for a few weeks - and is surely worth it in terms of increased productivity and decreased eRRORS (especially in a case-sensitive language).

Edward Collier

<>

As this issue's Student Code Critique Competition is a C one I hope you will be able to submit a critique. I do try to keep some balance, but, while finding bad code is not that difficult, finding it while trying to do a dozen other things means that it is likely to be whatever is to hand. I would be very happy for readers to submit code for critiquing. Note that I have to be careful about pulling code from books for this purpose.

I think that your point re Silas' article is covered in the previous letter. And basically you are right.

The problem with your suggestions for Quark Express is that I heavily use character styles (which, to my surprise, were only introduced to Quark Express in release 4). In addition, for all its faults, Word is a much better editor than Quark Express. Now that I have discovered how to export my Quark Express style sheets to Word (something that none of the PE's I worked with knew how to do) life is a lot easier. Now if Quark Express supported user defined hot-keys better and added a multi-level undo, I would be happier with importing straight ASCII to it. In summary, I find Quark Express fine for layouts but mediocre for formatting and bad as a word processor. Word is much better for formatting and very good as a word processor. However before Microsoft quote me, I should add that every version of Word I have ever worked with has time consuming bugs. One day that will kill me by inducing a heart attack.

Notes: 

More fields may be available via dynamicdata ..