Jump to content

Talk:Executable compression

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Lzexe

[edit]

The article should mention that "LZEXE" (written by some young kid in France nobody had ever heard of before) pretty much single-handedly invented this category of programs (at least as far as widespread use on microcomputers goes), leaving PKWARE and the others hurrying to catch up... AnonMoos 09:38, 15 August 2007 (UTC)[reply]

No, it did not. Fabrice Bellard wrote this program in 1989-1990 (according to his own website), at which time executable compressors had already been used routinely for several years by e.g. cracker and groups on the C-64.[1][2] --Viznut (talk) 11:26, 5 September 2009 (UTC)[reply]
Maybe -- but it still made a big splash in the MS-DOS world, and left much more professional and established software companies like PKWARE hurrying to catch up... AnonMoos (talk) 16:04, 18 December 2009 (UTC)[reply]

References

Merge

[edit]

Support

[edit]
  1. Yes, Exe Packers and Executable compression should be merged -- they are synonoms. The Exe Packers article seems much better maintained and descriptive than does this one. --db90h 08:44, 12 December 2005 (UTC)[reply]
  2. Support - Squilibob 07:24, 22 December 2005 (UTC)[reply]
  3. Merge. --DCrazy talk/contrib 14:20, 24 December 2005 (UTC)[reply]
  4. Merge -- Taral 08:30, 4 January 2006 (UTC)[reply]

Oppose

[edit]

I oppose merging the two, especially while the term 'packer' is The title of this page is "executable compression". Everywhere on the page this is described as making an executable programs smaller. Obfuscation is listed as a side effect.

A 'true' packer performs some kind of compression on the executable. Many of the tools listed in the "List of packers' do not actually compress the binary; some make it much larger than the orignal. While obfuscating sometimes happens as a side effect of compressing a program, packers generally are not designed with that as the primary goal.

If the intent of a tool is to make it hard to analyze, it is a protector or obfuscator, not a packer.

Packing is how you make something fit in a smaller space. Obfuscating is how you make things hard to understand. Protection is the goal of the obfuscation.

Unfortunately, the term 'packer' has grown in some circles to mean "any tool which modifies an executable so it doesn't look the same as the original."

VMProtect, for example, is not a compressed executable. VMProtect is a virtual machine based emulation of the original code, which takes much more memory, disk space, and CPU to execute. It is not possible to extract the original code from a VMProtect app.[1]

It's a protector, and it uses very strong obfuscation techniques to do this. It does not pack the program, so it is not a 'true' packer.

If anything, this page needs to make that distinction clearer.

Neutral

[edit]

Would you be so kind to mention also applications which are able to detect these often bad used exepacker/compressors ? For example, see at http://www.z80.eu/otherdelphi.html ... there is an application which is able to find these files in a generic way also. Please consider this. Thank you. PeterSmith123 (talk) 11:38, 15 May 2008 (UTC)[reply]

FSG

[edit]

The external hyperlink on FSG is not helpful. It leads to a page which seems not related to anything similar to FSG. —Preceding unsigned comment added by 217.86.142.7 (talk) 14:34, 2 September 2009 (UTC)[reply]

Good catch, link changed. --HamburgerRadio (talk) 17:43, 2 September 2009 (UTC)[reply]

PeSpin?

[edit]

Virus or not? — Preceding unsigned comment added by 75.70.114.202 (talk) 15:11, 6 November 2011 (UTC)[reply]

Directory of software

[edit]

Wikipedia is specifically WP:NOT supposed to be used as a software directly. Listings of nonnotable software do not belong here. It has been suggested such a listing is a 'valuable source of information', but many things are valuable (for example, a phone directory), but are nonetheless off mission for this project. - MrOllie (talk) 11:52, 8 April 2022 (UTC)[reply]

The article should include some mention of LZEXE by Fabrice Bellard, who was just a teenager in France when he played a major catalytic role in bringing executable compressors into use among ordinary end-users in the MS-DOS world... AnonMoos (talk) 19:18, 8 April 2022 (UTC)[reply]
Even Bellard's own biography doesn't include any details about that. Maybe start there? MrOllie (talk) 19:32, 8 April 2022 (UTC)[reply]
You came and wiped off 10 years of editing of this article with one swipe in name of I don't know what. Stop vandalising wikipedia and maybe focus on extenting this topic. I can provide 20 academic papers as a reference for every one of those exe packers listed. Winele8 (talk) 20:34, 13 April 2022 (UTC)[reply]
Sometimes stuff that doesn't meet policy persists for a while (sometimes a long while). That does not mean it can never be fixed. If you can provide twenty references each, I suggest you start writing an encyclopedia article for each - assuming that you're here to help build an encyclopedia and not to try to force in a link directory. Are you associated with one of these compressors? From your other edits it appears likely. Please keep in mind that Wikipedia policy requires such connections to be disclosed. MrOllie (talk) 22:07, 13 April 2022 (UTC)[reply]
This is ridiculous. This is an article about executable compression and executable compressors, and it is encyclopedically highly relevant and also perfectly normal in Wikipedia to include lists and tables of examples, in particular if they are historically relevant, define(d) a new or specific field, are technically interesting because of their particular way of how to achieve their task, or are important to include to give context for other entries or article contents. Wikipedia has hundred-thousands of lists, even dedicated list articles, and depending on the scope of a list, they also include entries which would not warrant an article on their own, in fact, including them in lists is often a way to avoid writing articles about each item. Also, while it is sometimes impossible to achieve completeness in a list due to the high number of potential entries, there is nothing wrong in aiming for completeness and perfection - actually, this is among the very goals of our project!
Ideally, each list entry should at least have a reference indicating that the entry does in fact exists. Depending on the list, it may (or may not) be desirable to include some extra information shortly describing the entry. Entries of particular importance may even be discussed in the prose or in articles of their own, but there is certainly no requirement for this.
Since you were invoking WP:NOT: You obviously misread that policy, as it does not advice against lists of software products but against "exhaustive logs of software updates", quite a different thing. There is nothing in WP:NOT which would disallow, or even discourage including lists of executable compressors in an article about this topic, in particular not the early ones used on various home computers and of CP/M, DOS, OS/2 and early Mac and Windows times - they were defining a field, were firsts of their kind in one way or another and are historically relevant - this is top-notch encyclopedic contents!
Yeah, we should work on further improving the sourcing and also provide references for those entries which currently lack sources - this has been an ongoing work for years and it will have to continue for years into the future due to the amount of work involved.
In the long run enough material might accumulate to discuss some of the particular imporant entries in more detail in the prose (it's not that it doesn't exist, more that it needs a knowledgable editor in the field - and time - to put the pieces together), but all this is long-term work in progress. Rome wasn't built in a day and we have WP:NODEADLINE. It is our normal editing process to collect bits of information over years until enough contents has been found to convert lists into prose.
Your bulk deletions of even carefully researched historically relevant and sourced contents left a previously medium-well-sourced article in a state with no references at all. This is a slap in the face of those editors who actually spent time and constructively contribute to articles, and it is harmful to the project and our collaborative editing environment.
I have therefore restored the last stable version before your deletions. If you want stuff to be deleted you will have to find consensus for this first on this talk page. I might agree that we cannot (nor should) maintain a complete list of current executable compressors and only list those which are special for some reason. But in regard to the older ones, they are historically and encyclopedically relevant, and they are few enough so we can (and should) aim for completeness. Perhaps we can find a better way to present this information mid-term. And let's find more sources.
--Matthiaspaul (talk) 11:01, 20 May 2022 (UTC)[reply]
This is a long list of nonnotable software packages. Much of it is completely unsourced, and what sourcing there is generally is not independent - it does not establish that anyone aside from the software authors have ever had a word to say about any of these programs. The lists inclusion is against the spirit and the letter of WP:NOT (specifically 'Wikipedia is not a directory') - it is not encyclopedic content. MrOllie (talk) 10:57, 18 July 2022 (UTC)[reply]
Stop your disruptive editing! Several editors have now asked you to stop removing historically, topically and encyclopedically relevant (and partially even well-sourced and highly accurate) contents! Your interpretation of WP:NOT is simply and fundamentally wrong: In an article about executable compression, lists of executable compressors are desirable and it would be against our goal as an encyclopedia to not mention at least those executable compressors which were firsts (or early) in an environment, provided some special features or properties, were well known at their time, or are interesting for some other reason (including to list them in context of other entries). There is no requirement for individual entries to have articles on there own or be sourced (in fact, providing them as list entries is often a desired way to avoid writing individual articles). However, a list certainly becomes more informative and useful for readers if the entries give at least a little bit of context.
Not all entries achieve this at present, but this does not mean you could bulk-delete the contents as a whole (and basically trash the article, leaving it without any references). It would not even entitle you to delete individual entries unless you know the information would be incorrect.
Since executable compression was a major achievement on 8-bit and 16-bit machines (because it had to be extremely small and fast to be useful, and there were no established ways to do it), I would consider all CP/M, DOS, OS/2 and early Mac and Windows executable compressors as well as all running on 8-bit home computers historically and topically highly relevant to be included in the article - and there are few enough of them to even attempt to achieve completeness without causing any bloat at all.
We should perhaps not aim for completeness for the Portable Executable compressors, because there are too many of them and on a 32-bit or even 64-bit machine executable compression is no longer much of a technical achievement. However, entries may be relevant for other reasons.
What I would like to see (and this is specifically allowed per WP:NOT is to see more entries enriched with information about when it was first introduced, who developed it, its magic signature (if any), the size of the decompressor engine, and other information. I would like to see more references as well, and I am adding them when I run into them, but that is a process for years to come.
--Matthiaspaul (talk) 20:18, 6 August 2022 (UTC)[reply]
Both WP:NOT and WP:Verifiability apply here; most of the entries here are completely unsourced, with no evidence of most of the entries here meeting WP:GNG notability criteria. OhNoitsJamie Talk 16:43, 24 September 2022 (UTC)[reply]