[openbeos] Newsletter #28

  • From: "Michael Phipps" <mphipps1@xxxxxxxxxxxxxxxx>
  • To: openbeos@xxxxxxxxxxxxx
  • Date: Thu, 17 Oct 2002 00:26:19 -0400

Semi-as-promised, here is the latest newsletter in email format, for those who 
won't read it unless it is in their inbox... This should be its own mailing 
list, soon...

The original, in lovely HTML is available at:   http://openbeos.org/nsl.php

Newsletter Issue: 28
# In This Issue: This newsletter and some administrivia by Michael Phipps
# Some vague ideas... by Michael Phipps
# Is it time for another BeBox? by Michael Phipps

This newsletter and some administrivia
by Michael Phipps [michaelphipps@xxxxxxxxxxxxxxxxxxxxx]

You may notice that this newsletter is a bit shorter than usual (and that I 
wrote it all). Remember, please, that unlike the Be, Inc. days, this is *your* 
newsletter as much as it is mine. We don't have paid staffers to 
coerce^H^H^H^H^H^H ask for articles. If you know something, share it. If you 
don't, ask on the public list for someone to write an article about it. Many 
people don't write because they don't know what to write. And don't worry if 
you think that you are not a good writer. We will work through it with you.

We are finishing up testing on the name voting system. Check back every day or 
so, and you will not be disappointed. Let's put this to bed...

I am slogging my way through legal processes on route to learning more than any 
human ever wanted to know about founding a non-profit organization.

Finally, as many of you may or may not know, I have been working diligently on 
a new virtual memory system. I would appreciate any and all feedback. You can 
find the code in the "vm2" directory. It is not a done deal, but a work in 
progress. Grab the source, read it, comment on it, test it, beat it up. It will 
make my work go much faster and help to ensure a fast, stable VM for all of us.

Some vague ideas...
by Michael Phipps [michaelphipps@xxxxxxxxxxxxxxxxxxxxx]

Some vague ideas...

Late at night, beating my brains for yet another newsletter article, my eyes 
were drawn to a notebook nearby my desk. This notebook contains my ideas that I 
brainstorm for future versions of OBOS. One of the complaints that people have 
is that R1 as we have defined it has no "future". It doesn't do anything beyond 
recreate what Be did and that we should aim higher for our first release. 
Interestingly enough, there is another, equal sized crowd of people who say 
that we are trying to do the impossible. In either case, R1 will be what it 
will be.

R2 and beyond are wide open, though. The following are some bullet items, some 
thoughts and ideas. Nothing solid, nothing developed, nothing thought through 
all the way, and certainly nothing promised. But some ideas, none the less...

XML - possibly methods for BMessage to transform to and from XML, possibly a 
whole different kit

Encryption - a plug in architecture for encryption/decryption. Probably 
public/private key. With a key manager of some sort, so that it is easy for 
users. Also with a tie-in to the Mail Kit of some sort so that mail programs 
can quickly and easily encrypt and decrypt messages. Also a cache of public 
keys with methods to refresh them (via Networking Kit)

Additions to BBitmap - this class is really in need of a lot more "stuff". 
Think LeadTools for Windows - the ability to choose some area of the bitmap and 
apply transforms (plug ins). This has been done a dozen times for a dozen 
different art programs. Let's do it once more, possibly leveraging someone 
else's work and make it part of the OS.

In a similar vein, how about a BSound class that adds effects? Mute, slide, 
reverb, echo, etc. All, again, as plug ins.

Vastly improved printing is needed. Color management is huge, now, with digital 
photography. Additionally, more functionality is needed. Watermarks, more 
margin control, etc. All of those "extras" that you have on Windows and use 
once a year.

BHTMLView - I am not sure how this one should be done, but it is just obvious 
to everyone that we need one.

Localization - this is probably the single most requested feature. And one of 
the biggest. Because this ties into so many other things. At minimum, we need 
to add currency, calendar and number widgets that tie in to the locale. Plus 
automatic translation support (i.e. translate phrases into current locale's 
language), resizeable, font independent widgets, right to left and top to 
bottom text, etc. There is a *huge* amount of work in doing this right.

Game Kit - this kit needs a massive overhaul. Be never really took it 
seriously, but I do. Sprites, Backgrounds, collision handling, parallax, 3d, 
scrolling, tiling, AI, controllers (joysticks, etc) and more. This kit could 
almost be a career for someone.

IM Kit - I have heard rumours that this is being worked on. Possibly leverage 
that into our work. This kit could be useful in a dozen different apps. Tech 
support in your office suite? Inter-player chatting in games? More and more, we 
are all connected...

Device Kit - if there is any less cared for kit out there, I don't know what it 
is. Only BJoystick (which belongs in Input Kit) and BSerialPort remain. But why 
does serial port deserve a class but not BParallel? Or BAudioCD? Or BVideoCD? 
Or Scanners, Cameras, PDAs, etc? Every "thing" that you plug into your BeOS 
machine should have a class to access it. Banish ioctl()'s forever!

Tracker - A lot of code clean up to happen here. After that, I would like to 
experiment with different, alternate Trackers. The desktop metaphor doesn't 
work for everyone. I don't think that anything works for everyone. Let's make a 
couple or more different Trackers and see if they suit someone's needs.

BClipBoard - ability to have multiple items on the clipboard

Interface Kit - there are dozens of widgets missing from here. Not to mention 
all of the stuff we mentioned for internationalization. How about an Outlook 
bar? Or a grid widget? Lots of honest work to be had, here.

Storage Kit - I think that this is a little hard to use. Lots of times, I have 
found myself taking the long way around on some simple things. Maybe some 
better methods...

Support Kit - Lots of new classes for here, too. There are numbers of useful 
little helper classes that we could all use... How about a pool allocator, as 
an example.

DataBase Kit - access for SQL databases will be a large sticking point to 
getting on the corporate desktop. The sooner the better.

Translation Kit - Good kit, as far as it goes. How about 3D objects? Or vectors 
(structured drawing)? HTML read/write? Tons of good work to be done here, too.

A CD Writer application is a critical piece that we lack. And a scanning app. 
Simple stuff that come with other OS's.

Power Management and other "special" laptop support

A dock - every OS should have one of these.

As you can see, there are a ton of ideas. And I know that every one of you has 
some, too. These are just mine. R1 is a stepping stone. A place to start. Not a 
place to end up. I certainly don't intend to rest on my laurels when R1 is 
done. I intend to step it up another notch and build what Be didn't have time 
to finish.
Is it time for another BeBox?
by Michael Phipps [michaelphipps@xxxxxxxxxxxxxxxxxxxxx]

Is it time for another BeBox?

Let me start off by saying that my introduction to BeOS started on a BeBox. DR8 
had just come out. A friend in college showed me his, then lent it to me for a 
couple of weeks. It was the greatest thing that I had ever seen. The lights, 
the cool bezel, the sheer blueness of it. The fact that it was a non-PC. It was 
awesome.

Be's published reason for eliminating the BeBox (and I have no reason to not 
believe them) was that it wasn't cost effective. That making dual processor 
machines was not worth the effort. Considering that Umax, Power and others were 
doing many of the same things and at very good prices, this seemed fairly 
reasonable. Be could concentrate its money and effort on building an OS and let 
others make the hardware. If they knew then what we know now... But they didn't 
and the past is over.

I have to wonder, though, if Be wasn't on to something that is more appropriate 
now than then. Times have changed. Hardware is drastically cheaper. And so 
phenominally faster that, at least in BeOS land, we haven't found applications 
to soak up all of that processor power. I know that my machine, even with only 
one Athlon 1600+ running (it is a dual processor), I would almost never 
saturate my processor. My machine always waits for me. I waste more CPU cycles 
in an hour than existed on earth in 1985. This is not an expensive machine, 
folks. Pricewatch.com shows my motherboard, 2 CPUs and 128 meg of DDR ram for 
less than $500, shipped. Memory, too, has far exceeded our average needs. The 
BeBox came with 16 meg of ram, if I remember correctly. Any more, 128 meg is 
the smallest size worth installing. Do I even need to mention hard drive space?

OK, well, so what? My PC is faster, cheaper and better than it was 6 years ago. 
No surprise. Moore's Law predicted this years ago. First of all, let's talk 
price. The BeBox was > $1000. And Be shipped them at a loss. Every one that 
went out cost them money. Today, one could create a machine that has 10X the 
power at half of the price and still make a (slim) profit.

Let's talk, very hypothetically, about what BeBox 2003 would be. The BeBox came 
from Be's NeoGeek period. Back when they made Be pocket protectors (yes, I 
still have mine). It was designed to be a machine to feed the inner geek. One 
tangible component of that market is that the machine doesn't have to make 
"realistic" sense. In the case of the BeBox, why would you spend $1200 or so on 
a machine that only runs an OS made by a start up company? Coolness. Straight 
coolness. So this machine would have to be cool to geeks. That means being 
technically better than anything else out there, as one could well argue that 
the BeBox was. It means having quality engineering and a clean, smooth look to 
it. Geeks and children instantly see through marketing...

The processor may well be the most controversial part. Many people will say 
that a BB2K3 *must* use PowerPC processors. I don't believe that. In the 
immortal words of the BeBox designer "it is dark inside the box". It really 
only makes sense to use the "best" processor out there, not the most 
fashionable. Be used PPCs because they were as fast or faster than x86 
machines, they had a clear future, they were fairly cheap and they had 
experience with them. For the hypothetical BB2K3, I would choose something 
exciting and low in price. The most exciting thing that I can think of right 
now, in the world of the microprocessor makers is the upcoming Clawhammer and 
Sledgehammer series from AMD. They are designed to be parallel processing, 
including HyperTransport for cross CPU (non-NUMA) architecture. Not to sound 
like an AMD marketing brochure, but that is really advanced stuff. IBM's 
upcoming PPC 970 sounds neatly designed, internally, but PPC is always in 
danger, anymore. Other t!
han Apple, it has no market. And with all of the "Apple is going to X86" 
rumours, well, I wouldn't bet the farm on PPC. Other chip technologies are more 
exotic and interesting, if a little long in the tooth - SPARC, MIPS and Alpha 
are among the chips that might make some sense to a company really interested 
in doing a lot of hardware work.

Another controversial decision that I would seriously consider is a more 
integrated motherboard. Yeah, I know. I like to think that I upgrade my PC, 
too. But face it - almost no one upgrades most of the pieces. I have the same 
sound, video, audio and networking cards that I started this machine with. 
Plus, there are cost, space and speed benefits to locating everything on the 
motherboard. I would *far* rather have a $200 motherboard with everything 
integrated than a $100 motherboard and $300 in cards to install. The $200 
motherboard will be smaller, easier to install and less likely to fail. Plus, 
with fixed hardware, drivers and configuration become a *lot* easier. Something 
that has always been a problem for this community.

One necessary component is the GeekPort. Yes, I think that it should come back. 
Absolutely. In fact, I think that there should be four of them. Let's get this 
thing used in scientific endeavors. There are lots of data collection 
opportunities waiting out there. I would build a USB based Geekport. Sell it as 
an add-on.

Similarly, das blinken lights would plug into the USB port. USB chips are not 
all that expensive. In fact, I would eliminate the PS/2 keyboard and mouse 
connectors. All USB, all the time.

An interesting concept was pointed out to me - solid state hard disks have come 
down in price. A long way down in price. And, interestingly enough, I would be 
willing to bet that one could put an entire, loadable OBOS R1 in a 64MB drive. 
What does that mean? How about blindingly fast boot time. Plus, all of the 
default applications start nearly instantly. Sounds almost like my Amiga days.

Form factor is also a critical component. As I noted above, the BeBox looked 
cool. I am the last person to design a cool case, but I certainly know them 
when I see them. As a Geek Machine, BB2K3 would have to be easy to get into and 
configure. It has to have smooth, clean lines. Little features of convenience 
are always a good thing. As an example - the Amiga 1000 was on legs so that you 
could slide the keyboard and mouse under it. $2 legs made an impression that 
lasted 10 years. I think that is a good tradeoff.

I doubt that the BB2K3 will ever get built. Today's venture capital market 
isn't exactly looking for high risk ventures to take on the Behemoth. But we 
can dream, can't we?



Other related posts: