Prev: DocFetcher - java-based desktop searcher
Next: alt.comp.freeware links at Fri Jan 15 21:20:02 2010
From: Caesar Romano on 14 Jan 2010 14:34
On Thu, 14 Jan 2010 06:28:45 -0800, mike <spamme0(a)go.com> wrote Re Re:
What Good is a Portable App that Won't Run?:

> Leaving behind JAVA seems excessive.
>If a program says it runs on XP, it should run on XP.

+1 on that.
--
Work is the curse of the drinking class.
From: Martin Jay on 14 Jan 2010 15:05
On Thu, 14 Jan 2010 10:25:22 -0800, mike <spamme0(a)go.com> wrote:
>Martin Jay wrote:

>>Doesn't it work with the portable version of Java:
>><http://portableapps.com/news/2009-11-30-_java_portable_6_update_17>?

>Thahks, I haven't looked very far into the .paf format.
>Will this run on any flash drive, or does the whole thing have to use
>the .paf paradigm?

It will probably run from any USB drive, and even from a directory on
a hard drive.

There are several USB portable suites around, but applications can
usually be mixed and matched between them.

U3 may be one notable exception. Despite having a U3 enabled drive I
haven't really looked at its functionality.
--
Martin Jay
Back the Ban: <http://www.backtheban.com/>
League Against Cruel Sports: <http://www.league.org.uk/>
From: JB on 14 Jan 2010 16:46
Yrrah wrote:

> JB <none(a)server.null>:
>
> > No, being "portable" really doesn't mean guaranteed to run on any
> > given machine. What it *does* mean is that on the machines where it
> > does run, it won't require an installation. Period.
>
> Nonsense. Period.
> http://www.portablefreeware.com/about.php

The nice thing about the Internet is that you can usually find an
official looking site that supports whatever you can dream up. I
prefer going with authoritative sources though, and oddly enough
portability does have an officially sanctioned definition. It's
covered by ISO 9126.

"Portability is the set of attributes that bear on the ability of
software to be transferred from one environment."

Note the glaring omission of "to another". ;)

So in actuality portability has nothing at all to do with running
on any arbitrary machine at all. It's 100% about the ability to transport.
To collect up all the bits in a nice neat package, for the purposes
of carrying it about. By definition that means not installing
anything outside a defined "container", be it a USB device, a CD,
or a separate partition or folder. So there's *the* definition of
software portability.

That put to rest...

Your own cite essentially agrees with me. That in the laypersons
world there's an "ideal" definition of portability... the pipe
dream... and that the dream itself is a moving target because your
dreams aren't mine. And then there is what's real. That there's all
sorts of requirements various users have, and that virtually none
of the "portable" software we're seeing today meets even the
authors narrowed down criteria, and remains useful.

I do like the author's breakdown of portability onto two general
categories though. Makes good sense, and also supports my position.
The author's "EXE portability" is 100% about not having to install.
Calling up an executable from it's native location without going
through any sort of moving of files or automagic system reconfigure
is the very definition of "non-install". Same same with settings
portability. Portable apps shouldn't even install ini/config
files. It's a clear breach of portability if you have to collect
up those bits manually when it's time to leave. Thank you. :)

From: JB on 14 Jan 2010 16:51
mike wrote:

> > Portability has nothing at all to do with not leaving traces. A
> > particular application *may* have that feature as a designed-in
> > goal, but it's absolutely not a prerequisite to portability.
> >
> Within limits. Leaving behind JAVA seems excessive.

Nothing is "leaving behind Java". You have a choice. You can run
the thing or not. If you run it you have to meet certain
requirements. They start with having access to a computer, and grow
ever more restrictive from there. If at any point you decide not to
meet said criteria you're free to vote with your feet. If you do
chose to meet them though, it's not an application leaving behind
remnants, it's you making a conscious decision to modify your
machine.

> If a program says it runs on XP, it should run on XP.
> Yes, the info says it's written in java. I'm not complaining about that.
> My only question was, "what good is a portable app that ain't portable?"

But the app you deemed non-portable was in theory more portable than
an app that might have been written in C or ASM, and ran on your
machine. More portable by a *huge* margin, given that it can run on
other platforms entirely.

> If it needs more than XP to run on XP, it shouldn't be called portable.
> Maybe we need a new term. How about "bloatable"?

I think you're thinking way to narrowly. You've convinced yourself
that because you ran into a system requirements issue there's a
problem.

I respectfully submit that you and your Win XP box are
insignificant specs in the universe, and that a pure Java
application approaches ultimate portability because it's OS
agnostic. And that that sort of broad coverage trumps a single
desktop and its update issues every time. :)

From: Spamblk on 15 Jan 2010 19:53
> I think that portable software should not require JAVA or .NET at all.
> But I never even download JAVA or .NET requiring software, 'portable'
> or not.
>
> Yrrah

Indeed. Though I once downloaded a couple of programs that then demanded
..Net. The programs were deleted without any further ado.


JB <none(a)server.null> wrote in
news:hin64v$tc9$1(a)news.eternal-september.org:

> mike wrote:
>
>> What Good is a Portable App that Won't Run?
>>
>> I just tried to run the portable version of docfetcher.
>> Sorry, Charlie, you need version blah, blah, blah of java.
<snip...>
>
> On vanilla installs of *which* OS? Mac, Windows, and Linux
> inclusive? Do you still support DOS 5 and Windows 3.11? 95/98? What
> about things like .NET? It's certainly part of the vanilla install
> of *some* modern, widely used operating systems. It's an additional
> installation to others. So does this mean portable applications
> can't be written using that framework? That would be odd, given
> the fact that part of the reason .NET exists is promoting
> portability (in a different but relevant sense). Ironically, Java
> is a similar tool. ;)

I don't think that "Stealth" (as Portablefreeware calls it) is strictly
required for portability but IMHO it is a very desirable attribute of
portability. Andrew of Portablefreeware even allows some programs that
modify the registry and accepts them albeit reluctantly as portable. Foxit
stores settings in the registry though it will start up happily without any
settings at all, IOW its at least no-install software.

I do not regard .Net as being compatible with portability though some
others do. There are computers using M$ XP in the local library system that
will NOT run .Net programs, therefore .Net I don't regard as portable. If
M$ decides .Net will be bundled with all Windows 7 machines and WIN 7
becomes as ubiqitous as XP is now then I would then think .Net does not
compromise portability.
First  |  Prev  |  Next  |  Last
Pages: 1 2 3 4
Prev: DocFetcher - java-based desktop searcher
Next: alt.comp.freeware links at Fri Jan 15 21:20:02 2010