I HAVE just been struck by the depressing thought that many of you may be reading this in the hope of finding out about the latest exciting developments in PC software. Unfortunately, there aren't any. `Exciting' is not a word which readily comes to mind when talking about programs for the PC.
Having had the dubious honour of using and reviewing dozens of databases, word processors, integrated packages and other types of software, I must admit that I am sick to death of seeing the same old thing regurgitated time after time after time.
I'm not entirely sure how to explain this sad state of affairs. Maybe the programmers are unable to think of anything better than has been done already. Or perhaps the manufacturers actively discourage original thought?
These days, when you get a spreadsheet, you know that it's going to come in the form of a matrix and that it will use formulas ‘hidden' in cells. But why should this matrix approach necessarily be the best way of storing mathematical data? Indeed, it could be argued that it is a highly inefficient way - most spreadsheets use so much memory that they can never approach anything like the 'maximum size' claimed by the manufacturer.
In any case, even small spreadsheets rapidly become unmanageable. It's difficult to find the figures you want and it can take ages to recalculate your work when even a trivial alteration is made.
One answer might be to hide figures in a hierarchical structure - a kind of spreadsheet equivalent of an 'ideas processor' such as PC Outline.
But companies don't seem to be interested in developing new approaches. If Lotus 1-2-3 is what people expect from a spreadsheet and Ashton-Tate's dBase III is what they imagine a database to be, then every other company is quite happy to go on churning out the same old thing endlessly.
Maybe we get what we deserve? But now that people are becoming more 'computer literate', let's hope that we'll all start demanding more - and not just out-of date ideas in fancy new wrappers.
It never ceases to amaze me how terrible most word processors are. An awful lot of them just go out of their way to make life difficult. Without wishing to venture too near to the laws of libel, I must say that I would be extremely unhappy to have to use even some of the `big name' programs such as WordStar and MultiMate on a regular basis. And I have a recurring nightmare that one day some particularly ruthless enemy (or employer) will chain me, kicking and screaming, to a PC and force me to use Samna Word!
As far as I am concerned, the only moderately-acceptable PC word processor is WordPerfect. It certainly has its faults (columns are awkward to set up, it can only use two windows and cannot have windows onto the same document...) - but in spite of that it usually doesn't stop me doing more or less what I want to. And it hasn't got all those unnecessary menus that seem to have become so unaccountably fashionable.
Until now, I considered the antithesis of WordPerfect to be Microsoft Word - lots of windows, endless menus, no macros - pretty but not actually very useful.
So I must say that the latest version of Word (release 4), comes as a welcome surprise. For those who like endless menus and useless boxes drawn around the screen, they are still there in abundance. But for the rest of us, it is now possible to remove them all and leave the screen virtually blank and looking, well, pleasantly like WordPerfect.
Microsoft has also added macros. And it has to be said that these are even better than the WordPerfect equivalent since they can be programmed using a macro language with control statements such as 'Repeat' and 'If ... Else'.
Still, I have yet to see a word processor that comes anywhere near the power and flexibility of Solution Systems' Brief programming editor (£185). This has a real macro language - one that lets you program completely-new features into the system, not just chain together existing ones in novel ways. It also has a proper 'undo stack' which can chain backwards through the last 300 editing actions.
I'd like to think that WordPerfect 5, which is soon to be released, will have all these facilities and more. But, somehow, I doubt it...
Words Fail Me!
One of the nicest features in Microsoft Word is its excellent thesaurus. This pops up in a window and lists all the synonyms of a word marked in the text. But, strangely enough, this is the one part of Word that was not developed by Microsoft itself. In fact, as the copyright notice tells you, it comes from a company called Microlytics.
This name might already be familiar to users of WordStar Professional, since Microlytics also provided the thesaurus for that program. And users of Symantec's Q/A database are given the option to buy a special Q/A thesaurus - not, of course, made by Symantec but by, you guessed it, good old Microlytics.
If you believe the spiel that the manufacturers give out about their products, a thesaurus is an `advanced' word-processing feature. And I guess an easy way of adding an advanced feature to a program is to buy one that already exists. In fact, the funny thing about all this is that virtually any word processor can have a Word Finder thesaurus because Microlytics sells a stand-alone pop-up version (£75 from In Touch).
All at C
At long last Microsoft's Quick C has arrived. The `Quick' of the name was beginning to sound like an elaborate joke since Quick C was supposed to be launched at the beginning of September whereas the first copies actually began to trickle through in mid-December.
The question is: was it worth waiting for? Well, I must say I haven't had the chance to do much programming in it yet. But first impressions are fairly favourable. It comes with a reasonable editor with the obligatory pull-down menu system and 'WordStar-compatible' commands. In fact, it is pretty much the same as Borland's Turbo C. Or is it...?
Microsoft would like to persuade us that Quick C is vastly superior to the Borland compiler and the company has even been putting out some pretty aggressive advertising point all the things that Quick but Turbo C hasn't. Top list is the Quick C debugger is like a simplified version of the Codeview Debugger provided with Microsoft C 5. ‘Breakpoints' and 'watchpoints' can be set to monitor variables during program execution while a highlight can be used to step through the source code itself.
The two other Quick C features that Microsoft is emphasising are the graphics library and I screen help for the C language.
As far as I can see, Microsoft’s claims for the superiority of the Quick C system seem reasonable. As far as its debugger goes I still prefer the C-Trace debugging that comes with the Mix C Works. Unfortunately Mix C can't offer much competition to Quick C or Turbo C because the Mix compiler itself isn't up to much - but I gather that a more powerful version can be expected soon.
Turbo C hasn't got any debugging, so in this respect Quick C has to win hands down in a comparison between the two systems. Personally, I prefer the Turbo C interface with its ability to collect together compiler warnings in a window and step through the source correcting them, one after another. And I find myself irritated by the typically Microsoft way in which Quick C's directory, search and help screens have selection boxes which must be tabbed to when using the keyboard (Microsoft like think that everybody will be using a mouse!). But worst of all, it drives me up the wall when the bloody thing warbles at me in such a self satisfied way whenever I make the wrong choice.
What I really don't undo is why Microsoft is selling two different compilers. While Quick C can be yours for £75, Microsoft’s C 5 costs all of £345. Does this mean that Quick C is deliberately crippled in some way to justify the extra £275 C 5 will cost you? I'd like to give you an opinion on the matter - unfortunately Microsoft C 5 ("available in September 1987") hasn't yet arrived in the UK at the time I am writing this ... (late December).
While on the subject of C, I am constantly amazed by the sheer brazenness of Zorland C's director, John Haggins. I mean, as if wasn't bad enough naming the company Zorland in the first place, Haggins now wishes us to believe that his recent renaming of the company to Zortech was merely "to avoid confusion".
In case, you hadn't realised exactly which company he wants to avoid confusion with, let me remind you of a certain compiler manufacturer called Borland which does a fairly brisk trade in Europe and has an even bigger operation on the other side of the Atlantic.
Now, why, after all this time, has Haggins suddenly been smitten with pangs of conscience? Well, it may not be entirely unconnected with the fact that Zorland... I mean Zortech, is currently launching itself in the USA. And, which product is it promoting over there? The Zorland C compiler, perhaps? Nope! Haggins reckons the Americans have had a gutsful of cheap C compilers and so he's launching a range of source code toolboxes - especially designed for Borland's Turbo C!
...Has the man no conscience?
Zorland/tech's sensitivity about its name may be well justified. American companies are easily upset by their names being taken in vain. A little while ago, the games company, Infocom, was obliged to change the name of its newsletter, The New Zork Times, after warnings from lawyers acting on behalf of a rather better known journal coincidentally called The New York Times. I must admit to being quite amused at the thought of The New York Times worrying that millions of potential sales were being lost by readers mistakenly buying the Infocom rag.
The nice people at Infocom have just sent me a copy of their latest game, Beyond Zork. This looks very promising, having been written by Brian Moriarty, the man responsible for the best ever Infocom adventure, Trinity. Beyond Zork actually draws the maps on the screen for you, which seems a remarkably civilised thing to do. I confess that, being something of a Zorkian purist, I am rather suspicious of the addition of a lot of 'roleplaying' stuff though. Beyond Zork insists that you set your character's `attributes'. Surely, if it was a really clever game, it should be able to work those out for itself. I think I'd probably get a high score for Ruthless, Low and Cunning. And Lazy - I usually try to let somebody else type in all the commands to a game while 1 sit back in a comfortable chair having occasional brilliant ideas...
I've been playing about with a new collection of utilities called File Rescue Plus recently. I say `playing', though Heaven knows, utilities like this aren't normally the kinds of thing you use for the fun of it.
Basically, File Rescue Plus is the latest alternative to the Peter Norton Utilities. Peter Norton was the bloke who had the clever idea of reading all the IBM and Microsoft documentation that nobody else could be bothered trying to decipher. Then, using all the `magic numbers' lurking around in the operating system, he put together a set of utilities which appeared to be able to do the impossible such as bringing a deleted file back into existence.
File Rescue Plus does all the familiar Norton tricks, but it has one very big difference - the price. Whereas Norton Utilities will set you back £85, File Rescue Plus costs a very reasonable £24.95.
In addition to its file 'unerase' facility, File Rescue Plus has plenty of other good things. It can be used as a friendly 'front end' to DOS (very trendy!), allowing you to navigate the directories of a hard disk with ease, it can locate 'lost' files by searching through directories for a specified file name, with or without wildcards. It can draw a `picture' of a disk showing which files are fragmented and where bad sectors or cross-links occur. It can even 'unfragment' files, putting together pieces which have become scattered about the disk, thereby increasing the speed of program loading.
As far as I can tell, File Rescue Plus does all it claims to do and does it efficiently. The trouble with any program of this type is that it is virtually impossible for a reviewer to give a true picture of how good the utilities are since the most useful ones can only really be put to the test when disaster has struck. And I have no intention of deliberately corrupting my hard disk just to prove a point!
Personally, I think I shall stick to PC Tools 3 (£30). This is a bigger set of utilities than File Rescue Plus. It even includes a hard disk backup program. And, what's more, it can be used as a pop-up so that the PC Tools utilities can be used even when you are not at the DOS prompt. I'm told that PC Tools Deluxe is on the way. I'll let you know more about that when I get a copy.
When Computer Shopper was launched in the UK the only colour was to be found on its cover. The contents were entirely monochrome, mainly text with very few pictures, printed on cheap and nasty off-white paper. It cost 78p and contained about 130 pages. The main news items were that Atari and Commodore were battling for dominance in the home computer market and that a new version of the Amiga computer was under development. Acorn meanwhile was investing in Unix as the operating system for its Archimedes computer and Amstrad had sold more PCs in the UK during 1987 than IBM.
Apart from the Rants and Raves column, I see that I also wrote a fair whack of the rest of the magazine including a three page comparative review of Amstrad's hideous but cheap PPC just-about-portable PC (£399) and Toshiba's less hideous but expensive T1000 (£995). I wrote a 1 page on the Modula-2 programming language and a two page review of Borland's Turbo Pascal 4 compiler.
Looking through the adverts, I see that hardware and software was amazingly expensive back then. A basic dot-matrix Epson printer would have set you back £205 (about $386 ), a cheap laser printer cost about £1,200 ($2,263 ), a 20 megabyte hard disk cost £215 ($405 ) and a Compaq PC with a 286 processor, 640K of memory and 20Mb hard disk would have set you back something in excess of £2,575 ($4,855 )!
...Interestingly, my musings upon the need for a new type of spreadsheet which didn’t hide formulas in cells and had a more ‘hierarchical’ view of data preceded Lotus’s wonderful Improv spreadsheet by a couple of years. Improv was exactly what I’d been hoping for. It used English-language type formulas and let you display cross-referenced ‘multi-dimensional’ views of data (this was later copied by - I mean ‘provided the inspiration for’ - the pivot tables in Microsoft’s Excel). Improv was initially launched on Steve Jobs’ innovative but unsuccessful NeXT platform and was subsequently ported to Windows. However, Lotus’s long established spreadsheet, Lotus 1-2-3, was already dominant on the PC and one suspects that the company did not want Improv to compete with its own top selling product. So they threw away the better spreadsheet in favour of supporting tired old 1-2-3. This left the way open for Microsoft to take the lead with its slightly less tired old Excel. As a result, the spreadsheet today is much the same as it was in 1988. Dull.
...You have to remember that the PC world was still using MS-DOS operating system back in 1988. No multi-tasking, no mouse control and graphics - no Windows! The dominant application software did not come from Microsoft. The top selling word processor was WordPerfect, the top database was Ashton-Tate’s dBase and the top spreadsheet was Lotus 1-2-3. There was far more competition among software companies in those days so that, for example, a substantial number of people used word processors such as WordStar, MultiMate, Volkswriter and Samna Word. Most of these products were absolutely terrible. MultiMate contained some of the worst bugs it has ever been my misfortune to be bitten by - and Samna Word was just hideous. Word processors back then did not have fancy stuff such as multi-level undo and few lf them could even display more than one document at a time.
Microsoft wasn’t having all its way in the world of programming back then either. Previously, PC programming was dominated by big, expensive and unfriendly C compilers such as Lattice C. By the late ‘80s, however, various companies such as Borland, Mix and Zortech had begun to introduce small, friendly, inexpensive C compilers. Microsoft tried to take them on at their own game with the launch of Quick C. Later, Microsoft even tried to compete with Borland’s popular Turbo Pascal compiler with the launch of Quick Pascal. Neither Quick C nor Quick Pascal took the world by storm, however. Which just goes to show that Microsoft doesn’t always get it right….