| |
Calum Tsang wrote:
> The two laptops I brought over were Thinkpad T23's which are 1.14Ghz P3Mobiles
> with 256MB of RAM each. If that's not enough for Java, I don't know what is.
> In case my memory serves me incorrectly, at worst, they were T21's, which are
> 800Mhz P3 Mobiles.
Oh, forgot about trying out those laptops. But then the biggest problem
we had with those was freeing up the serial port from multiple levels of
security, networking and anti-virus software that was installed on them.
I'm surprised anyone could any work done on them. The battery life
would be exhausted after the 3 hours it takes them to boot.
But really we're still talking serial ports here. Something the mac
hasn't even had for at least 5 years probably 10. So you example is
still based on antiquated hardware. I'm taking about the realities of
today. Not 1998.
> The problem with such a complaint is that the developer always argues the
> existing deployment base is outdated, while the market reality is that very few
> people will have brand new 2.8Ghz PCs with 1GB of RAM. The market will always
> struggle to keep up and you can't demand that everyone upgrade. It's even worse
> with Macintoshes, due to Apple's performance gap (real or imagined) and the fact
> Mac users tend to keep their hardware for longer.
With every release of OSX, the OS gets smaller and faster. The
performance of older equipment is actually improving with time.
At no point have I suggested a hardware performance level. Other then
it's got to be realistic. I was thinking those 486 compac laptops. I
think it's reasonable in the vast majority of cases to drop those off
the supported list.
The factors for setting hardware requirements are entirely depending on
what your trying to do. There is a quite reasonable install base of
both Windows and Mac hardware to handle software that, for example, Lego
is writing Today.
> As for who uses the serial port....well, in the Lego context I can think
> of...Dacta Control Lab, the Mindstorms IR tower, CyberMaster's RF tower, hrm,
> what else...Spybotics...jesus, that's half the frickin product line, isn't it...
Again I'm talking Today, not 5 years ago. Dacta Control Lab - not a
public product, no longer available from Lego. CyberMaster - no longer
available from Lego, Spybotics - no longer available from Lego.
Mindstorms - only available with USB tower.
So the only hardware product left that is still being sold by Lego is
Mindstorms, and it's USB. Perfectly compatible with the majority of
today Macs and PCs.
> First, except for corporate applications, Java UI's, from this former UI
> designer's perspective, are absolute crap. They provide poor feedback and break
> UI consistency conventions with the OS. I honestly can't see anyone developing
> a good crossplatform UI, especially for children, in Java. Ugh.
One, I haven't suggested a technology. Just because I work with Java,
doesn't mean I think it's the solution to every problem. There are a
multitude of solutions out there. You have to consider what is the
right solution for the job.
Two you clearly haven't done anything with Java in the last 5 years.
Just like with any programing language, you can make crap and you can
make art. With Java I find it's reasonable easy to do either one.
> Second, in my opinion, the hard part of VisionCommand is the fact it has to
> capture, process and act on USB video streams. If you feel it's easy to code
> for completely different capture architectures and act on that data, well, I'm
> no programmer. You're probably right then. All I have to support this is a
> beta trial of a nonlinear editing system that I worked on, where they never
> solved QT versus VFW architectural and performance issues. Strangely enough,
> the package was also written in...you guessed it...Java. That and using the
> first five versions of Premiere for Windows, another QT ported video
> application, dog slow and unresponsive.
The demands of a nonlinear editing system are way beyond that of
VisionCommand. However I would be surprised if Java was the best
solution, or at least the exclusive solution to either of these problems.
> Well, your original complaint was about Lego products on the Macintosh. Lego's
> computer based products so far have broken into:
>
> -Mindstorms like devices (which need serial, USB)
Mindstorms like devices (which need USB) They haven't been serial for a
couple of years. Next generation release of Mindstorms, I'd like to see
Mac support.
> -Games like Lego Island (which need DX or some sort of graphics engine)
Some of these games could easily be cross platform, some not.
> The former, you've already written off as inconsequential, the latter you agree
> to. So really, the cross platform issue in the Lego context is nontrivial.
> We're not porting some database front end here for accountants...the Lego
> products hit the one thing cross platform projects always get snagged on:
> Hardware.
You're forgetting the website. There's no reason the Mosaic application
couldn't be cross platform for example. Since I don't have my Mac yet,
I don't know what else might be an issue on the website.
> But yes, the PC hardware issue is a big one. That's more of a "Why you should
> own a Mac" argument though.
And then there would be more demand for cross platform support. :-)
> > > Then you've got to support the thing once it hits market. Which means call
> > > centers which have Mac AND PC experience. Maintaining a pool of Macintosh
> > > support operators with additional training.
> >
> > Separate training, not additional training.
>
>
> So when I staff my call center, and no one calls with Mac problems, I should
> keep these staff sitting around idle while there are other inbound calls? I
> don't think you've had the pleasure of supporting a product.
Good point. Either way it's not really a big deal. This is where good
product design will save you a lot of effort. If the product works
similarly on each platform then you need to train people on the product,
then have special training for install and quarks of the OS.
I work on a product that hasn't had proper customer support for 5 years.
I know what it's like to support a product. 25% or more of my time is
taken up by it.
I also get to watch the products that get proper support and see how
well the process works when done right.
From a cost perspective it's stupid the way the are doing things with
my product. But because of quirks in budgeting it doesn't hurt the
bottom line of the right people so it will never get fixed.
> > Think about it 5% of the market is a hell of a lot of people. There is
> > money out there to be had.
> > However I did manage to get you to wine about how hard cross platform
> > development is. :-)
>
>
> No, I'm just offering a slice of reality to the typical fanatic or technically
> minded, who never seem to realize that there are much more difficult issues to
> answer than the most basic technical concerns. People who go off and claim that
> 5% of the market is a lot of people, without backing it with real numbers.
> People who would demand Mac support without understanding the cost of
> development or support staff. Unless you're willing to support it with a
> business case, my general answer to Mac and Linux users is to shut up and live
> with it. It's the same as those who complained about grey colour changes or
> that Lego doesn't care about AFOLs as a market segment.
In reality I agree with this. My only desirer is that when a company is
developing a product that they make the business case that it's not
worth doing, rather then just listen to people spout off that it's not
worth doing without a business case to back it up.
The reason to draw attention to it is to make people think about it the
next time the decision to support another platform comes up. Rather
then dismissing it out of hand.
The complaining about the color issue when it first came up was quite
legitimate. Lego created a significant problem for AFOLs. The
complains made Lego go back and look at the issue. They realized they
left out a small but significant portion of their customer base when
they made the decision to change the colours.
Although they ultimately decided it didn't make business sense to go
back, they did make the effort to review the situation.
What's annoying is the people who are now technically wining about it,
now that the decision has been made.
From my perspective, this is all I hope for: If you going to support
Mac or any other platform have a good reason for it. But conversely if
your not going to support a platform also have a good reason for it.
Derek
|
|
Message has 2 Replies:
 | | Re: iMac
|
| In lugnet.org.ca.rtltoronto, Derek Raycraft wrote: <snip> (...) Our very own o.t-d :) So I work for a major school bus transporation company as a 'tech support' guy. And, having 500+ systems in the field, occasionally we will get the odd "Imminent (...) (20 years ago, 21-Oct-04, to lugnet.org.ca.rtltoronto)
| | | Re: iMac
|
| (...) As I noted in (URL) some of the software Lego has outsourced was to a company called Qube, using a cross platform "Q Engine" that runs on Windows, Linux, X-Box and PS2 platforms. Check out the old post for the links. Pretty cool stuff for (...) (20 years ago, 22-Oct-04, to lugnet.org.ca.rtltoronto)
|
Message is in Reply To:
| | Re: iMac
|
| (...) The two laptops I brought over were Thinkpad T23's which are 1.14Ghz P3Mobiles with 256MB of RAM each. If that's not enough for Java, I don't know what is. In case my memory serves me incorrectly, at worst, they were T21's, which are 800Mhz P3 (...) (20 years ago, 20-Oct-04, to lugnet.org.ca.rtltoronto)
|
26 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
