Re: mr on Apple web site

Robin Barclay wrote:

>Well .... some of us PC afficianados hate Macs just as much - personally I
>have hated them since they first made it difficult to access their DOS and
>write programs in any language (especially for accessing/controlling lab
>apparatus) - back in the '70's when there were several different options -
>not just PC's.  I am glad that the "PC" became a standard (there were too
>many diferent systems) and that IBM did not hang on to it the way Apple kept
>the Mac in house (you get more for your money with a PC because many
>different people make them).  You don't often get PC people knocking Macs
>the way that the Mac people knock PCs - there seems to be a lot of Microsoft
>paranoia.  In my opinion PCs are much more common and versatile in labs than
>Macs - especially outside the USA - and will eventually become the standard
>for interfacing with lab equipment..... and they can "look cool" if you shop
>around for the right case if that's important to you.

I can't pass up the opportunity to get into the PC/Mac battle - I hate them 
both, but there aren't really alternatives (yes, I know, there's Linux, but 
it's probably even harder to hook up a Linux system to hardware than it is 
to hook a Mac up to hardware).

Of course, what started this thread was the assertion in the piece about 
Mario on Apple's web site that there were 15,000 to 20,000 Mac-based flow 
cytometers out there. The best estimates I had from industry people in 
early 2003 when I was finishing up the 4th Edition of PFC was that the 
total number of systems from all manufacturers in use was under 20,000. If 
that's correct, BD would need a 75% market share to account for the low end 
estimate of 15,000 Mac-based machines. I'd be interested to know where the 
15,000 to 20,000 figure came from.

The old (68000 series and possibly early PowerPC) Macs were difficult, but 
not impossible to connect to hardware; it was easier to work with the PC's 
ISA bus, which, while slow, was perfectly adequate to do most flow 
cytometric data acquisition and analysis. There were decent versions of 
Forth, which was one of the first and best languages designed for 
controlling hardware from mini- and microcomputers, for both PC (DOS) and 
the Mac (Forth was the first Mac programming language made commercially 
available, at a time when the only other option was buying the very 
expensive Lisa from Apple on which to develop Mac software). I found 
MacForth easier to program with than the Windows Forths (or other Windows 
languages, e.g., Delphi), but my old Macs used to crash all the time.

So what makes me unhappy with both Windows [and Windows machines] and the 
Mac in their current incarnations? It is now significantly harder for mere 
mortals to write software to get hardware to communicate with either PCs or 
Macs; the gain in complexity associated with the PCI bus, USB/USB2, 
FireWire(IEEE 1394), etc. is greater than the gain in speed and 
convenience. Also, in making the operating systems more stable (and Win XP, 
despite its security issues, is almost as stable as Mac's Unix-based OS X), 
both Microsoft and Apple elected to eliminate the ability of their computer 
hardware to respond rapidly to interrupts (latencies are now in the tens of 
thousands of instruction cycles), meaning that any really fast hardware 
attachment for either a PC or a Mac now needs to have a DSP in it, whereas 
if the fast interrupt response had been preserved, the hardware attachments 
could have been made much simpler and cheaper. Linux also takes fast 
interrupt response off the table, so it doesn't represent a viable 
alternative. If you go to Apple's web site and look at what data 
acquisition hardware is available for Macs, particularly for Macs running 
OS X, there isn't much, and many of the companies that supported the Mac in 
that area have dropped Mac support for their newer products. That is 
undoubtedly one big reason why BD's newer digital pulse processing 
cytometers are running on the PC platform.

That doesn't stop anybody from analyzing FCS files on Macs. FlowJo is 
well-conceived flow cytometry software; one reason it is as good as it is 
is that it was written by people who do a lot of flow cytometry, and 
cutting edge flow cytometry at that. But there are other people who do a 
lot of good flow cytometry who have written good software, for PCs as well 
as for Macs.

For the record, I have a G4 PowerBook, which I use mostly for the iLife 
applications, which are slick. It doesn't crash more than once every couple 
of months, but the same is true of my Windows 2000 and XP systems. OS X can 
be as infuriating as Windows when one or another aspect of it goes counter 
to your intuition or to what you have gotten used to. Macs, while somewhat 
more expensive, are much better made than many PCs, and they are certainly 
aesthetically pleasing. If there were a reasonable alternative to 
Microsoft's Office applications for the Mac, I might consider switching. 
The 12.1" PowerBook is a nice portable, but it's over a pound heavier than 
the Fujitsu laptop I now use, which has pretty much the same speed, memory, 
and drive capabilities (OK, not DVD-R, but I don't burn a lot of DVDs).

However, I really wish Apple had stuck with the plan they had a few years 
back of writing a Mac operating system for Intel hardware. I think that 
died when Microsoft bailed Apple out with a few hundred mil. If it were 
possible to run OS X and XP mano a mano on the same hardware, there'd be a 
more rational basis for comparison. But, as may be the case for the 
November election, minds, once made up, are not easily changed.

-Howard