What Wikki thinks about portable software.

Discuss anything related to portable freeware here.
Post Reply
Message
Author
User avatar
tactictoe
Posts: 283
Joined: Thu Dec 10, 2015 10:56 am
Location: A galaxy far far downunder
Contact:

What Wikki thinks about portable software.

#1 Post by tactictoe »

Wikkipedia has a list of portable software (Incomplete), but more interesting is the debate or academic debate it did generate due to the creation of the previous page.
This debate can be read here https://en.wikipedia.org/wiki/Talk:List ... e_software. If you are just in interested in the list or curious why the debate start check the list they propose https://en.wikipedia.org/wiki/List_of_portable_software. I know anyone can edit Wikki page or add on but, hmmmm it makes you ponder why the page exists in the first place. Also some honorable reference to this web site exist on the 'Talk:List of portable software' mentioned earlier in this post. An interresting point found on this TALK was the definition Wikki gives to portable. Unfortunately and once again it is NOT the majority that win here as it was define by one individual only? I loved this one: What is the difference between Portable software and Software portable? Anyway, it worth a visit. And for us developer it is just blurry what is really portable and what is not as the definition of portable software is NOT regulated by any ISO. Should it be? So, to launch an academic debate and as an attempt to reach a good definition of what is a real Portable software, what is your definition?

As I fire this debate, IMHO here is mine:
- Should not be writing anything (as in settings, registry key, or whatever) to any media without user will.
- Should be adaptive with its path, and should run from anywhere without affecting the software works unless specific options saved by the user of this software.
- Should be able to clean up any mess it creates if any, by mess I mean temp file and the like and this mess should be located into the portable software folder somewhere.
- Should be stable on any platform it is supposed to work


You turn, have a nice day. :D

User avatar
SYSTEM
Posts: 2041
Joined: Sat Jul 31, 2010 1:19 am
Location: Helsinki, Finland

Re: What Wikki thinks about portable software.

#2 Post by SYSTEM »

tactictoe wrote:Wikkipedia has a list of portable software (Incomplete)
Wikipedia, with one K.
tactictoe wrote: but more interesting is the debate or academic debate it did generate due to the creation of the previous page.
This debate can be read here https://en.wikipedia.org/wiki/Talk:List ... e_software.
As you can see, they have had multiple debates (there is a section for each discussion). In addition, all of the debates are very old: the last one was held in 2007.
tactictoe wrote: If you are just in interested in the list or curious why the debate start check the list they propose https://en.wikipedia.org/wiki/List_of_portable_software. I know anyone can edit Wikki page or add on but, hmmmm it makes you ponder why the page exists in the first place.
Yes, it indeed doesn't make much sense for Wikipedia to have a list of portable software. There is so much portable software out there that it's best to leave cataloging of portable software to websites which specialize on it (out of which we are the largest :P ). We have features such as search, categories, extraction instructions and accurate portability information such as the "Stealth" field. Wikipedia doesn't have any of that.

In fact, the article was nominated for deletion. Twice.
tactictoe wrote: An interresting point found on this TALK was the definition Wikki gives to portable.
Wikipedia. Also, don't abbreviate "Wikipedia" as "wiki". Wiki is a form of a website (see https://en.wikipedia.org/wiki/Wiki). Many other wikis exist, such as Wiktionary (owned by Wikimedia, like Wikipedia is) and PCGamingWiki.
tactictoe wrote: Unfortunately and once again it is NOT the majority that win here as it was define by one individual only?
Wikipedia tries hard not to define things on their own; they try to find external definitions. Besides, there isn't anything wrong with the definition they're using. It mostly matches our definition of stealth.
tactictoe wrote: And for us developer it is just blurry what is really portable and what is not as the definition of portable software is NOT regulated by any ISO. Should it be?
Even if there was an official definition of portable software, I think that not many people would care.
tactictoe wrote: So, to launch an academic debate and as an attempt to reach a good definition of what is a real Portable software, what is your definition?

As I fire this debate, IMHO here is mine:
- Should not be writing anything (as in settings, registry key, or whatever) to any media without user will.
- Should be adaptive with its path, and should run from anywhere without affecting the software works unless specific options saved by the user of this software.
- Should be able to clean up any mess it creates if any, by mess I mean temp file and the like and this mess should be located into the portable software folder somewhere.
- Should be stable on any platform it is supposed to work


You turn, have a nice day. :D
My definition is this:
  • Should save settings in program folder, or alternatively, not have settings at all.
  • Should not have dependencies which are unlikely to be fulfilled (too new version of Windows, too new version of .NET, a third-party library/runtime which most computers don't have [e.g. Python], etc.)
My definition is quite permissive.

Definitions don't matter too much for us, though. Thanks to the voting system, our definition of "portable" is essentially "anything that gets enough votes in the database". :)
My YouTube channel | Release date of my 13th playlist: August 24, 2020

User avatar
tactictoe
Posts: 283
Joined: Thu Dec 10, 2015 10:56 am
Location: A galaxy far far downunder
Contact:

Re: What Wikki thinks about portable software.

#3 Post by tactictoe »

About the Wikki thingy, I am Aussie and tend to short word all the time like any Aussie. Wikki or Wiki is Aussie slang. But true enough it is confusing and should have been more careful to not use any 'local' slang into an international forum, so point taken, no offense taken. and thank you for remembering me :D

:arrow: Vote system is probably the best method but it will frustrate some developers.
However votes brings something very good on this site:
- People are genuinely testing and creatively doing critics/suggestions of submitted software.
- Dialog is open between people and the developer/author.
In that respect: a software evolve faster, idea are created to improve the software... all in one, it's all for the best.

:arrow: I don't think ISO or any regulation on the definition will change the way developer works. when .Net was born Microsoft made it very clear to all developers: 'You are with us, or against us'. Too bad for the company in Redmond, it did not work as well as they hoped. IMHO, I believe anybody defines 'Portable', it's own way regarding needs of the moment. Still interesting to know what are the common factors of all these definitions.
My definition is this:

Should save settings in program folder, or alternatively, not have settings at all.
Should not have dependencies which are unlikely to be fulfilled (too new version of Windows, too new version of .NET, a third-party library/runtime which most computers don't have [e.g. Python], etc.)
I like your definition, clear and precise. :wink:

Dependencies are difficult/impossible to avoid, unless you include in your code as resource dll and the like inside which will probably violate copyright if you redistribute non-owned material. On the other hand creating a 'plagiarism' or clone of these OS dependencies are tasks nearly close to the impossible too for practical reason. Your code will also be a monster in weight, and windows evolve to fast to keep up (like you said) to be able to maintain such type of code. But you already knew that, :wink:

Have a nice day. :D

User avatar
webfork
Posts: 10818
Joined: Wed Apr 11, 2007 8:06 pm
Location: US, Texas
Contact:

Re: What Wikki thinks about portable software.

#4 Post by webfork »

How about less of a definition and more of a sliding scale?

User avatar
tactictoe
Posts: 283
Joined: Thu Dec 10, 2015 10:56 am
Location: A galaxy far far downunder
Contact:

Re: What Wikki thinks about portable software.

#5 Post by tactictoe »

Sliding Scale system sounds good, as it is a rigid system but revisable. Definition is relative to the point of view of an individual or its perception of the definition: too much factor per individual to be united here. A scale is mathematics and therefor more easily accepted. However it is better as a rating system, and may be a way to achieve a good definition of portable software. A scale permits also automation of process to define if an application is portable or not, and is probably far more flexible when it comes to revision.

Your Scale (which is actually imho a logical break down definition of portable software -> not portable to totally portable):
Will break if it's run outside Program Files
Writes settings to the registry / appdata folders
Doesn't require additional installation/downloads (exceptions are made for Visual C libraries, dotNET, and Java, but their use isn't preferred)
Writes settings to the application folder
Stealth (writes nothing anywhere besides the app folder or temp folders)

A software could do these test easy but point two (some serious code to be done there)... hmmm you just gave me an idea for a new test software now, :-)
I guess there are as many definition of portable software as they are humans trying to define it. Which give us a potential of 7.4 billion and growing potential definition.
Let's keep the scale idea open, it breaks the definition into simple clear point everyone can agree. So Scale it is instead of definition.
Thank you for the idea and keeping this conversation alive.
:D

User avatar
Midas
Posts: 6705
Joined: Mon Dec 07, 2009 7:09 am
Location: Sol3

Re: What Wikki thinks about portable software.

#6 Post by Midas »

Been wanting to chime in on the debate here, but didn't make it up to now. Let me start by endorsing webfork sliding scale approach, I'm all for non-absolutist stances.

Moving forward and picking up from SYSTEM's proposed definition, my portable concept reads thus:
  • So long as program integrity requirements are declared and met, moving a program location (computer/drive/folder) should never break functionality, nor cause loss of settings; naturally said requirements should be minimized and easy to attain, hence the preference to saving settings to program folder or the valued ability to recognize relative paths and system variables.
  • On the subject of dependencies, keeping in line with the previous principle, the absolute minimum should be expected -- e.g., no DotNET for programs designed for Windows XP, where it was an add-on, while in later Windows versions DotNET v3.0 is reasonable as it's included in a default installation. Regarding library and runtimes requirements, these should also be kept portable whenever possible -- e.g., meaning DLLs should work if present in the program folder or another convenient relative location customizable through settings. As a further example, please note how Java apps are made portable with Jportable.
Lastly, as I have repeatedly declared here, for me portability always trumps stealthiness, no matter how convenient the later is. If the worst comes to pass, one can always recourse to launchers and system cleaners... :oops:

P.S.: check how a thoughtful developer expertly dealt with some thorny limitations at http://www.petges.lu/home/support/#pane ... c-portable...

Post Reply