Jump to content

User talk:Fabio Maria De Francesco: Difference between revisions

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Content deleted Content added
Linux Kernel: Update.
Tags: Mobile edit Mobile web edit
Linux Kernel: Blob examples.
Tags: Mobile edit Mobile web edit
Line 24: Line 24:


[[Special:Contributions/185.89.35.4|185.89.35.4]] ([[User talk:185.89.35.4|talk]]) 21:12, 31 July 2020 (UTC)
[[Special:Contributions/185.89.35.4|185.89.35.4]] ([[User talk:185.89.35.4|talk]]) 21:12, 31 July 2020 (UTC)

Here is one such example of a blob:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/net/wan/wanxlfw.inc_shipped?h=v5.8-rc7

Here is another:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/scsi/aic7xxx/aic79xx_seq.h_shipped?h=v5.8-rc7

Ans here is one more:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/gt215.fuc3.h?h=v5.8-rc7

As I'm sure you can see, these blobs are long sequences of bytes disguised as .c source files. No C programmer can understand them effectively. They are proprietary, binary blobs.

Thank you. :)

[[Special:Contributions/185.89.35.4|185.89.35.4]] ([[User talk:185.89.35.4|talk]]) 21:39, 31 July 2020 (UTC)


== System programming language ==
== System programming language ==

Revision as of 21:39, 31 July 2020

This user is alternative name of fabio_m_de_francesco. Fabio Maria De Francesco (talk) 20:53, 15 December 2015 (UTC)[reply]

Linux Kernel

Hello, I'd like to thank you for all your edits to the Linux kernel page. You are obviously very knowledgeable about the kernel.

However, I must ask if you know about Linux-libre, as the deblobbing efforts are ongoing and quite real. I run Parabola with a variant of the Linux-libre kernel, and it obviously has an effect because my laptop's internal Wi-Fi and Bluetooth card will not work due to missing firmware; proprietary firmware which Linux-libre removed from the mainline Linux kernel. I also get many "/* DEBLOBBED */" messages on boot, which indicate blobs were trying to be loaded, but they were removed; proprietary microcode, for instance.

Blobs in the Linux kernel are not obvious. They are hidden and sometimes well-hidden in plain site. For instance, some binaries are in .c files as long arrays of bytes. I invite you to download the Linux-libre source tree to read the deblob script, and compare the deblobbed source tree to the mainline one.

These binaries are not GNU GPLv2-covered, and would be violating the GNU GPLv2 if they were, because they are not in the "preferred form of the work for making modifications to it".

Thank you. :)

185.89.35.4 (talk) 19:24, 31 July 2020 (UTC)[reply]

I just read what you wrote on Talk:Linux kernel#Linux kernel licensing rules.

May I ask why you show disdain by stating that contributors who disagree with you are not even worthy of the term (by calling them ""contributors"")? Please can you show proof that Linux-libre is just "GNU propaganda"? Please can you explain why Linux-libre exists if the mainline Linux kernel is blob-free?

Have you seen what the Linux-libre deblob scripts remove and check for? Have you seen that they are indeed blobs? Do you want me to point you in the direction of one so you can see for yourself?

Thank you. :)

185.89.35.4 (talk) 21:12, 31 July 2020 (UTC)[reply]

Here is one such example of a blob: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/net/wan/wanxlfw.inc_shipped?h=v5.8-rc7

Here is another: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/scsi/aic7xxx/aic79xx_seq.h_shipped?h=v5.8-rc7

Ans here is one more: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/gt215.fuc3.h?h=v5.8-rc7

As I'm sure you can see, these blobs are long sequences of bytes disguised as .c source files. No C programmer can understand them effectively. They are proprietary, binary blobs.

Thank you. :)

185.89.35.4 (talk) 21:39, 31 July 2020 (UTC)[reply]

System programming language

Normally assemblers, compilers, linkers, etc. are classified as systems programs. Peter Flass (talk) 03:35, 23 June 2019 (UTC)[reply]

@Peter Flass:

First of all... I thank you. I know you have a lot of experience, mostly with IBM machines that I've never seen. My first computer was an dismissed S/36, my second one an AS/400. When you started in 1967 I was 1 year old.
That said, I still think that the set of system programs contains just what is needed for computers to bootstrap themselves, to provide consistent interfaces to users while hiding the intricacies of hardware, to create and kill user processes, to make arbitration of time and space resources, to evaluate credentials and authorizations of user applications...
In short: system programs comprise all and only what is needed for computers to run user applications. Not more, not less.
Obviously, different architectures require different sets of system software. Hypervisors, might be needed or not. HPC's running hundreds of thousands of distributed nodes require MPI and other middleware that mainstream PC's don't need.
Let me return to compilers, debuggers, assemblers, and the like... for sure they are made of very low level code. Compiler developers know their targets at a deeper level than most kernel developers. Notwithstanding compilers aren't needed in order to run other user applications.
Furthermore compilers run in user space only if system programs retrieve load and assign resources to them. If the operating system kills processes running compilers, all other processes wouldn't even notice or care Fabio Maria De Francesco (talk) 22:50, 27 June 2019 (UTC)[reply]

Moving from a user's talk page to the relevant article's talk page

One of my edits was reverted by a very experienced user. We have exchanged some messages in his/her talk page but we still disagree. I'd like to go on to the relevant article's talk page in order to know what other users have to say on the subject. Obviously I'd like that these users could read what was reverted and see all the messages already exchanged. How can I solve my issue? Thanks in advance Fabio Maria De Francesco (talk) 23:03, 11 November 2019 (UTC)[reply]

Add a note to the article's talk page that the issue already has been discussed at User talk:DVdm#Reverted edit in "Mass" article (with the link). Regarding the reverted changes, a link to this diff may help. Huon (talk) 23:42, 11 November 2019 (UTC)[reply]

Page numbers

I added {{Page needed}} to each of the book citations that you recently added to Logic. If you can add specific page number ranges to the citations (with |pages=), that would be helpful. Thanks, Biogeographist (talk) 13:26, 25 December 2019 (UTC)[reply]

@Biogeographist:
Following your suggestion, I added information about the relevant pages for citations #2 and #3, but not for #4. Please let me explain...
Citation #2 comes from a specific page, therefore it was an easy task.
Citations #3 and #4, in my opinion, can only be verified by carefully comparing most (if not all) the chapters in Quine and McGinn's books; Although I think that chapters 1 and 6 of Quine's might suffice to support verifiability, conversely I'm not able to select any particular chapter from McGinn's book.
Notwithstanding the Wikipedia's guidelines state that page references are optional, I remain open to any suggestion if it can help in solving my difficulties with this issue. Greetings, Fabio Maria De Francesco (talk) 18:40, 26 December 2019 (UTC)[reply]
Thanks for adding the page numbers! I interpret Wikipedia's guidelines as saying that page numbers are required unless a statement refers to a book or article as a whole (or to an unpaginated source, of course). Wikipedia:Citing sources § Identifying parts of a source: "When citing lengthy sources, you should identify which part of a source is being cited." Biogeographist (talk) 19:01, 26 December 2019 (UTC)[reply]
@Biogeographist:
As I wrote in another page, unfortunately, I love English but Italian is my mother tongue; at the same time I'm a newbie here at Wikipedia. This is why I'm sure that your interpretation of the guidelines must be correct. As soon as possible I'll try to select ranges of pages from the McGinn's book too. I own that book and I read it a lot of time ago, but believe me when I say it'll be a hard work. However I'll try and do my best. Thanks for your time and kindness. Fabio Maria De Francesco (talk) 19:25, 26 December 2019 (UTC)[reply]
@Biogeographist:
I've added a link to a review of the book of McGinn that I had cited. That review explicitly compare the different points of view between McGinn and Quine about the scope of Logic. I hope it will be sufficient for supporting my edit. As I said it's very difficult for me to provide the pages you requested, because I think that the book as a whole should be cited. Please, if you have time, read the above-mentioned source and let me know if the template "page needed" can be removed or not. Thanks

Edit warring

Regarding [1] and [2], please have a look at wp:FIRSTSENTENCE, wp:edit warring and wp:BRD. You can discuss on the article talk page. Thank you. - DVdm (talk) 09:50, 19 January 2020 (UTC)[reply]