Re: OS X.4.X update...



The Translucent Amoebae <transamoebae@xxxxxxxxxx> wrote:

The Languages thing sounds like that might take up some space, But i
think that the user should be able to choose/control which languages
are being supported for any given computer.

You can, at the time you install Mac OS X. Sometimes updates will "put
back" some languages you have removed, unless they are smart enough not
to install missing languages.

The Intel version of the 10.4.11 delta update is 127.4 MB compressed
(with an extra layer of compression for the disk image or software
update download, to take it down to about 109 MB) and it expands out to
348.6 MB.

The vast bulk of the 10.4.11 update is /System (209 MB for Intel) and
most of that has no language-specific elements, or if it does they are
relatively tiny (kilobytes).

The second biggest lump is /usr (hidden Unix stuff), which is 72 MB, and
has very little in the way of multiple language support.

The third biggest lump is /Applications, which is 48.4 MB. Nearly all of
that is Safari 3 (almost 40 MB), and of that about 28 MB is language
support files.

In total, I'd estimate that language support files make up less than 10%
of the total size of the 10.4.11 update. If you were able to get an
English-only version of the update, it might save something like 8% of
the file size. In order to do this, Apple would have to supply about
sixteen different language-specific versions of the update, which would
be a vast increase in the amount of work for them to generate and
distribute the updates, as well as disk space on their web servers to
store all of them.

Multiply this by the number of alternative versions of the update (delta
PPC, delta Intel, combo PPC, combo Intel), double it again for the
Server edition of Mac OS X, and double it again for the Software Update
copies as well as the manually downloadable versions, plus some extra
variants through Software Update for model-specific updates (which are
cut down versions of the delta updates for recently introduced models).

With multi-language updates, I'd estimate that they need about 18
different files available for the 10.4.11 update, totalling about 2.8 GB
on their servers.

If they did separate indivdual language updates, the total storage
requirements on their servers would be more like 40 GB (which has to be
replicated on all nodes of the Akemai network), and it would only save
you about 8% of the download time.

They'd also need the multi-language version so that anyone who actually
did want more than one user interface language didn't have to do two or
more downloads.

It's just crazy for them to choose them for me. ( principally because they
will invariably leave out some family dialect that most bilingual users
need... Using the "popular" languages would only be appropriate for a
small subclass of business users...??? )

The language support is primarily the user interface aspect of
applications, and allows the sale of computers in countries which speak
those language as their primary tongue or at least as a widespread
secondary tongue.

The've covered the major Western European languages, Russian, Chinese,
Japanese and Korean. Others are being added regularly (e.g. Polish made
it into Leopard). I expect more will be added in future as Apple expands
into more markets.

You can write text in a much wider range of languages (in modern
applications which support Unicode - AppleWorks is not one of them).

But aside from that; i grew up ( so to speak ) on a OS7.2 Performa and
the biggest application that i had then was ClarisWorks, which was 5
or 6 separate applications that all worked together, and it did just
about everything... and it was only 4 MB.

Just about everything for a simple office/productivity application, but
with major shortfalls in some areas (addressed by more advanced packages
in each of the areas supported by ClarisWorks/AppleWorks).

It doesn't even touch on the other things you can do with computers now
which were impossible in the early 1990s. Digital photography comes to
mind as a major application.

So i will go to my grave believing that there is a conspiracy to
convince users that they have to keep buying newer and newer computers
to run fatter and fatter ( severely bloated ) software...!

I agree that software bloat is a problem.

The issue is that as CPUs have got more powerful, and memory and hard
drives have increased in capacity, and video hardware has improved in
resolution, it has become possible to do a lot more with computers,
which means more and bigger applications to support all the features a
user might want. It also gives the software developers the excuse that
they need to use all this extra capacity.

A fair amount of "bloat" is due to eye candy: the Mac originally used
16x16 icons, but Mac OS X went up to 128x128 and now Leopard supports
512x512, which gives an incredible level of detail if you enlarge an
icon, but I don't see that it serves a particularly useful purpose, and
results in icons occupying about 16 times as much disk space.

Changing processor families also introduces increased code size, due to
having to maintain two sets of code for compatibility with old computers
and old applications (e.g. the 68K to PowerPC transition, and the
current PowerPC to Intel one).

The entire world of computing has got more complicated, with the
introduction of an ever increasing set of file formats, file systems,
network protocols and connection methods, web site features, and
peripherals which can be connected to a computer, which requires an ever
increasing set of software to deal with all the variations.

If there was a small and reasonable incrementation to memory, to
accomidate larger graphics files, and if they wanted to come up with
faster computers, that would be fine...
But to make it ABSOLUTELY NECESSARY for EVERYONE that wants to use
Email and the Internet to Keep Upgrading to entirely new computers,
and to have to keep changing formats for everything, including simple
text documents, just so other people can read them is wrong.

E-mail is one application where it is still possible to use a very old
computer and get a reasonable experience (as long as you avoid HTML or
formatted e-mail and don't want modern file attachments).

HTML and the web in general is a moving target, so if you want to be
able to access a wide variety of web sites and content you are generally
forced to keep your web browser and plugins reasonably up to date.

File formats are a problem. One of my pet hates is the habit that some
people have of distributing files in proprietary formats such as
Microsoft Word or Excel, assuming everyone else with a computer has paid
for the appropriate version of Microsoft Office. The use of widely
supported interchange formats like PDF, RTF or HTML (and the growing use
of XML) allows a greater degree of interoperability and backward
compatibility.

If you stick with an old computer and operating system you then run into
the problem that you are eventually in a tiny minority of the market
share, and major software developers don't tend to update software for
people who refuse to buy newer software, operating systems and/or
computers.

The latest versions of applications often want to make use of features
which require the latest operating system, or which might work with
reduced capacity in the previous operating system, but if you are two or
more major versions of out of date, you will rapidly lose options in
terms of new software, or updates for the old software you are still
using.

Performance is also an issue - some applications are simply impossible
on old computers, because the CPU isn't fast enough.

There should have been a standardization of elemental file formats,
that would accommodate newer features being introduced every few
years, established 30 years ago, that would still work today...

Agreed. Plain text files are reasonably well standardised (a few minor
variations such as line endings) but issues like supporting languages
other than English require newer standards such as Unicode replacing
ASCII.

We are gradually getting there in some areas. For example, PDF has been
a reasonably well established standard for many years and even as new
features are added it is possible to at least get a partial view of the
document in older tools. RTF is similar. Graphics file formats are also
reasonably well standardised (GIF, JPEG, TIFF, etc.)

I hope that ODF takes off, at least as a well defined interchange format
for Office documents as well as a "native format" for OpenOffice and
derivatives, plus new office software.

The user that bought a computer 30 years ago, to use as a word
processor, should still be able to use that old commadore, and with
slight upgrades to memory and a plug in board, connect to the
internet.

8-bit computers with 128 KB of memory (or less) simply don't have the
resources to use the Internet in any practical sense. About the best
they can achieve is text-only applications like E-mail. They don't have
enough processing power or data throughput to cope with graphics to the
extent that modern computers can. In many cases they don't have the
capacity for large scale data storage or signficant amounts of
additional memory, due to the limitations of technology available at the
time, and may not have any external interfaces which are capable of
using Ethernet, limiting them to a dial-up modem connection or similar
(and then probably at slow speeds by recent standards).

Interface technology has moved on to newer and better standards such as
USB instead of ADB and serial ports. They allow significantly faster
performance and improved feature sets such as automatic device
identification and hot-swapping, but this comes at the cost of requiring
a much more powerful computer to manage the USB bus.

Some development are due to economies of scale. For example, SCSI is a
much more powerful interface than ATA for hard drives, but ATA was more
popular and cost less to produce, so it became the mainstream format,
limiting use of SCSI to high-end applications. Old Macs which only
support SCSI are basically impossible to maintain, because hard drives
eventually wear out and there is no affordable way to get a new hard
drive for the old computer.

It shouldn't be necessary to have to buy an entirely new computer
every 6 months, just so you can use email.

That's a ridiculous claim, even for more rapidly moving applications
than e-mail.

My "replacement cycle" for old computers has typically been in the order
of five years, occasionally faster if driven by major technology changes
(e.g. I bought an iMac G3 for USB/Firewire/Mac OS X), or because I want
a different type of computer (e.g. laptop for portability), or because I
wanted a more powerful computer for a new class of applications (e.g. a
G4 for digital photography), or a second computer for special
applications.

Of all the computers I've bought, only my Apple IIgs is unable to
support e-mail as an application (and that could be resolved with some
software to add TCP/IP support, probably not to a degree I'd be
satisfied with). You could still do e-mail on a contemporary 20 year old
Mac SE with 4 MB of RAM running System 7.1.

Web browsing is fine on any G4 or later Mac, only slightly limited on a
G3 Mac, and still feasible with limitations on older PowerPC and 68K
Macs dating back to the early to mid 1990s.

Applications involving heavy number crunching, broadcast quality video
or image manipulation (such as digital photography) are only pratical on
G4s and later models.

Advanced graphics and high defintion video are only practical on Intel,
G5 or high-end G4 models.

--
David Empson
dempson@xxxxxxxxxxxxx
.



Relevant Pages

  • Re: Meyer(s) gets it wrong
    ... but not in the same way that many, supposed, OO languages support them. ... I have two programs running on seperate computers. ... TBH I find the idea of seperate processes which have defined interfaces, ... be described as being 'just' good SP practice. ...
    (comp.programming)
  • Re: Normal areas between two values: a home-made program
    ... Because the algorithm was do to span values greater than 0, the different signals of interval bounds were took into consideration. ... of computer hardware including process control computers. ... I've written code in other languages including C and C++. ... interpreter ...
    (sci.stat.math)
  • Re: advantages of forth over other languages
    ... but the people who actually accomplish things with computers have other priorities. ... The purpose of a programming language is to express a solution within the paradigms, abstractions, models, and idioms supported by that language. ... "learn new languages" means understanding different ways to conceptualize and create a solution for a problem. ...
    (comp.lang.forth)
  • Re: Python syntax in Lisp and Scheme
    ... > But what if computers someday become equally capable ... > contrained by the form of I/O and that other languages, ... > on speech, free-form 2D writing, or other forms of input may ... As for lisp, that's ...
    (comp.lang.python)
  • Re: using Linksys "gaming adapter" w/5 port workgroup switch
    ... >I want to set up internet access for a several computers in a different room ... >from my wireless gateway (Linksys WAG54G). ... As usual Linksys support is mostly correct. ...
    (alt.internet.wireless)