--- Forwarded Message from Charles Oliver Wolff <[log in to unmask]> ---

>Date: Wed, 10 Feb 1999 09:50:27 -0500
>To: Language Learning and Technology International Information Forum               <[log in to unmask]>
>From: Charles Oliver Wolff <[log in to unmask]>
>Subject: Re: #4845.2 Any labs using iMacs? (!)
>In-Reply-To: <[log in to unmask]>

------------------ Not meaning to start a debate over PC's vs. Macs. That's not the issue here. The issue is how robust is an iMac. After looking inside, I'm suspicious. And I've been looking inside Macs, PC's, Unix boxes, and all sorts of custom hardware for twenty-three years. 
>
>Regarding upgrading, this concern is vastly overstated here as elsewhere.  How many labs actually upgrade their
>logic boards?  

Well, Noyes Lodge at Cornell does. With the price of components as low as they are these days, it the cheapest way to go with a PC if you have the talent in your department to do so. On the Mac side of things, it is true that it makes more sense to replace the whole machine unless catastrophic failure of a motherboard in a waranteed machine makes that unnecessary. That is a best case scenario. But in general, all over our campus, there is a lot of old hardware still in use. I've swapped and replaced a lot of internal components on both platforms, including motherboards in Macs, just to keep people going. 

>
>The iMac IS a serious lab machine. Who else, for example, has the guts to get rid of that useless floppy? 

As a tech with more than 20 years of experience in academic and corporate settings, I will agree that floppies are nearly useless for the end-user these days. But they sure do make support an awful lot easier. One of those things you never miss until you need it. If I use one even once a year to get a machine back on its feet, its paid for itself. 

 I manage
>50 lab Macs and have responsibility for another 25 faculty and staff Macs.  I can't tell you the last time I needed
>floppies or got software on them.  That's what a network is for. 

You're certainly right about that, that is what a network is for. But in the real world, errant machines can't always see that network. Well, maybe that necesary utility is on a CD somewhere, providing our CD extension isn't corrupt. Or sometimes we remember some little used, nearly forgotten tool that just might do the trick on a shelf somewhere. But...doggone it...it's on a floppy. That's the end of that! 

The iMac was made for the network (LAN or
>Internet).  And who else makes a machine of such performance with such a small footprint?  Component systems would
>be too cumbersome in our carrels  and would probably require modifications to the furniture (we currently have the
>one-piece Power Mac 5260s).

Again, I agree with you. But if you remember back to the first time Apple integrated a monitor and CPU in a single case, you'll also remember why they became unpopular. Here we are, a couple of years down the road, and the monitor dies in our iMac. No longer under warranty. Now what do we do? Installing a new display into that machine will be a lot more costly than just plugging in a new one. The problem sounds like component systems are not an option in YOUR lab environment. That's a reasonable statement. But that does not mean that, in general, component systems end up being more economical from a service and support point of view, even component systems with a higher initial cost. Whenever these kinds of systems are marketed, they soon fade away for these reasons. Remember the old Kaypros? Or the first Compaqs? As well as those old Mac's, the service cost was outrageous and sooner or later, a useful internal component was scrapped along with a failed unit...the baby with the bath water again. How soon we forget. 

>
>With regard to having to replace the hardware soon in a lab with heavy traffic, well, we have about 500 students a
>semester using 50 lab Macs nine hours a day.  In three years I have replaced one hard drive.  Headphones, yes,
>plenty of those.  But is the implication here that PCs are more durable than iMacs or Macs in general?  Puhleeze.

No. That's not the implication. That's your interpretation. The fact of the matter is whether you are using a Mac or PC, the standard expected useful life of a desktop machine is three years. Not just because of the stress on the hardware in a heavy-use environment, but also because that's about how long it takes software to outpace current capabilities of entire systems. Of course, we all face budget constraints and squeeze as much as we can beyond that, at least until complaints about inoperability, poor performance, etc. get routed to where the money is. We do that by upgrading components, something that an iMac is not going to be very friendly about. 

Not so long ago, I had faculty driving me nuts demanding I get Netscape to run full force on old XT's and PS2 386's. And some were even working on a bunch of old Mac SE's. Did I recommend they all buy PC's? Of course not. Support means giving the customer what they want. Because of those budget issues, we've had to keep a lot of old hardware running, and we'd replaced a LOT of motherboards, drives, and internal components in both PC's and Macs. 

>
>G3 instead of iMacs?  Sure, great idea--if you have the option of spending twice as much and have the room for them.

Now this is the real debate, here. With an iMac coming in at around $1,200 it is hard to beat. But a 300 MHz G3 is currently $1,600 at MacWarehouse. Hardly twice as much! Add the availability of educational discounts and clearances, etc., and a G3 looks better and better. Of course, if you have to make major modifications to your facility to accomodate them, well, a G3 is simply not an option as the savings fades away. So why do you ask? 

It's the long range cost I worry about. I once watched a huge faculty debate center around the cost of virus protection. They decided to go without it to save some money, even though the cost of labor for dealing with virus issues when they arose far exceeded the $10 per machine we would have paid up front, just as we had argued. But we're all working in a setting where in the money in the hand that enables or disables our decisions. 
>
>
>We stream RealAudio files to all of our clients from our NT with no streaming server software.  Ditto with MPEG
>files.

You don't need server software to view MPEG files across a network, but you are not really "streaming" them. That's a technical issues, though. If you are delivering Real encoded audio and/or video, and are not using a RealServer, you are creating an enormous performance hit on your NT server. And that, is simply a fact. You are not really "streaming" them, either. Whatever works, though...that's the name of the game. 

It will be very interesting to see how this all looks three or four years down the road. 




Charles Wolff

*** ZONE FORECAST ***

NYZ015-022>025-055-PAZ038-101600-
BRADFORD-CHEMUNG-SCHUYLER-STEUBEN-TIOGA NY-TOMPKINS-YATES-
INCLUDING THE CITIES OF...CORNING...ELMIRA...ITHACA...OWEGO...
PENN YAN...TOWANDA...WATKINS GLEN
350 AM EST WED FEB 10 1999

.TODAY...CLOUDY IN THE EARLY MORNING...THEN BECOMING MOSTLY
SUNNY. HIGH AROUND 40. NORTHWEST WIND 10 TO 20 MPH.
.TONIGHT...MOSTLY CLEAR. LOW 20 TO 25. LIGHT NORTHWEST WIND
BECOMING SOUTH.
.THURSDAY...MOSTLY SUNNY AND MILDER. HIGH IN THE MID 50S.