Re: [BRLTTY] The brltty service gets in the way of other serial port applications
Hello, Jason J.G. White, le ven. 15 déc. 2023 11:09:41 -0500, a ecrit: > On 15/12/23 10:23, John wrote: > > Secondly, disabling the service ought to have been sufficient to prevent > > the software from disrupting the operation of the system, but it wasn't. > > It had to be completely removed. Is this a bug? If so, where do I report > > it? I see the 'issues' function on the Github repository is not enabled. > Perhaps your Linux distribution's bug tracker, if it's ultimately a > packaging issue. Yes. Distributions are not supposed to install by default the very-generic-udev-rules files, so as to avoid disturbing generic usb/serial cards. But they should indeed install the braille-specific-udev-rules files by default, so that blind users can use them without having to ask somebody to configure anything. Samuel ___ This message was sent via the BRLTTY mailing list. To post a message, send an e-mail to: BRLTTY@brltty.app For general information, go to: http://brltty.app/mailman/listinfo/brltty
Re: [BRLTTY] The brltty service gets in the way of other serial port applications
On 15/12/23 13:53, John wrote: Dave, Thank you for your reply, which I found both enlightening and informative. This will also help me explain to others. In addition, you are more likely to encounter this issue if you have an old version of BRLTTY installed. As I recall, the developers tightened the udev rules at some point. For reference, version 6.6 is current. If you have a significantly older version, it's a matter for your Linux distribution's package maintainers. ___ This message was sent via the BRLTTY mailing list. To post a message, send an e-mail to: BRLTTY@brltty.app For general information, go to: http://brltty.app/mailman/listinfo/brltty
Re: [BRLTTY] The brltty service gets in the way of other serial port applications
Dave, Thank you for your reply, which I found both enlightening and informative. This will also help me explain to others. On 15/12/2023 17:04, Dave Mielke wrote: Disabling the brltty service doesn't prevent it from being automatically started by a USB udev rule. If you really don't want a Systemd unit to start then you need to systemctl mask it. I haven't heard of masking a service. To me disabled, means disabled, i.e no longer active or able to affect the system. However, having read up on it I see it is a 'stronger' version of disable. Thank you for pointing that out. I hope it's okay for me to point out that Arduino and friends are just as much to blame as brltty is. Brltty's problem is that some braille device manufacturers don't bother customizing their USB vendor and product IDs. Yes, they should do it. Also, however, so should Arduino and friends be doing exactly the same thing. There'll always be such conflicts as long as whomever doesn't stop using generic vendor/product IDs. I am wanting to better understand so its no problem at all. As far as I know, Arduino does customize its vendor and product IDs. Genuine boards should be appropriately identified. The problem is that there are a multitude of clone boards around, mostly from Far East sources which use generic UART chips and generic IDs. There are also boards from Adafruit, Sparkfun, STM and others. Not sure how good they are at customizing their product IDs. There are also many products from other vendors that have generic chips so I guess this is always going to be a problem. I had imagined that there ought to be some way of managing access to devices in udev or by detecting known Braille devices, but with all those devices with generic IDs about I can now see that this is not so straightforward. Can you suggest a way to design it that way? Since you expect it to be that way, you must have some idea regarding how to do it that way. A braille user would be unable to make such a choice without first being able to read his/her screen, which means, yes, that his/her braille deivce must already be working. Let's call it one of those "catch 22" situations. You make a good point here which I failed to appreciate. I would not imagine Braille user being able to install a Linux system for various reasons, for example not being able to see the BIOS messages or knowing when the system has actually finished booting, but if having brltty built into the distro makes that possible, then I absolutely support that. I had also imagined that if a Braille user can open, read and edit a document, that they ought to be able to respond to a prompt, but how could they do so if the software that enables this is not yet in place? It makes sense to have it already right there, in the distro. I guess everyone else just needs to be aware of it and learn how to disable it appropriately when its not needed. If it's earlier than 6.5 then your distribution is too old. As of brltty-6.5, we split our USB rules into two - generic and customized. So, all you really need to do is to ensure that you're at least at brltty-6.5 and then to uninstall the generic USB rules package. It looks like it is earlier than 6.5. I re-installed brltty from the repository and ran the brltty --version command as suggested and it returned version 6.4. My Linux Mint distro is up to date, but Mint is always behind Ubuntu/Debian in terms of the software package versions in the repository. I am therefore not surprised. -- John. ___ This message was sent via the BRLTTY mailing list. To post a message, send an e-mail to: BRLTTY@brltty.app For general information, go to: http://brltty.app/mailman/listinfo/brltty
Re: [BRLTTY] The brltty service gets in the way of other serial port applications
[quoted lines by John on 2023/12/15 at 15:23 +] >As per advice, I disabled both services, but this unfortunately did not >restore normality. In fact, the service was still being loaded as could be >evidenced by running the 'ps -ef' command to show a list of running processes. Disabling the brltty service doesn't prevent it from being automatically started by a USB udev rule. If you really don't want a Systemd unit to start then you need to systemctl mask it. >I should point out that Arduino is not the only area of electronics that I >work with, that also requires access to serial ports. I hope it's okay for me to point out that Arduino and friends are just as much to blame as brltty is. Brltty's problem is that some braille device manufacturers don't bother customizing their USB vendor and product IDs. Yes, they should do it. Also, however, so should Arduino and friends be doing exactly the same thing. There'll always be such conflicts as long as whomever doesn't stop using generic vendor/product IDs. >Whereas I appreciate the necessity of having such software for those who >need Braille, I don't understand why it was installed automatically as a >mainstream service as part of an OS update? So that blind people who use braille can actually read their screens. Try closing your eyes, promising not to peak even once, and interact with your system. It won't matter how well your monitor is working - you won't be able to do it. Are you suggesting that blind people should only ever install a Linux system when they have a sighted assistant? >I would expect such a program to be installed on an as-needs basis? Can you suggest a way to design it that way? Since you expect it to be that way, you must have some idea regarding how to do it that way. A braille user would be unable to make such a choice without first being able to read his/her screen, which means, yes, that his/her braille deivce must already be working. Let's call it one of those "catch 22" situations. >Secondly, disabling the service ought to have been sufficient to prevent >the software from disrupting the operation of the system, but it wasn't. That's how Systemd units work. An equivalent, for example, would be how a service unit is still automatically started if its corresponding socket unit is activated. >It had to be completely removed. Is this a bug? If so, where do I report >it? I see the 'issues' function on the Github repository is not enabled. No, it's not a bug. But, to answer your question, this inndeed is where to report bugs. >Thirdly, can the software be configured to ignore certain ports? Brltty tries as hard as it can to only consider actual braille devices. As mentioned above, however, we can't help it when some manufacturers choose to use generic IDs and we also don't wish to exercise any control or influence over which braille device any given user chooses to use. Again, however, I'll ask why is it that Arduino feels that it's okay for it to be using generic IDs. Does something make Arduino more important than a braille device? Of course not! It's all a matter of personal perspective, and, to me, that makes it wrong to favour one over the other. >Finally, if a program or update is likely to disrupt what might >generally be considered 'normal' operation of the system (I "generally" >because I accept that for users of Braille, having brltty installed is >their 'normal'), shouldn't the software at least warn the user of this >and request confirmation before installing itself? Again, how would you design it that way? How can a braille user read such a warning and/or respond to such a confirmation before his/her braille device is working? Now that I've griped a bit at you, I'll tell you the simplest way to deal with this issue. First, do: brltty --version If it's earlier than 6.5 then your distribution is too old. As of brltty-6.5, we split our USB rules into two - generic and customized. So, all you really need to do is to ensure that you're at least at brltty-6.5 and then to uninstall the generic USB rules package. >I am curious as to how and why it suddenly appeared on my system. Probably due to an upgrade to a newer brltty release which introduced support for a braille device that's using the same generic IDs as you are. -- I believe the Bible to be the very Word of God: http://Mielke.cc/bible/ Dave Mielke| 2213 Fox Crescent | WebHome: http://Mielke.cc/ EMail: d...@mielke.cc | Ottawa, Ontario | Twitter: @Dave_Mielke Phone: +1 613 726 0014 | Canada K2A 1H7 | ___ This message was sent via the BRLTTY mailing list. To post a message, send an e-mail to: BRLTTY@brltty.app For general information, go to: http://brltty.app/mailman/listinfo/brltty
Re: [BRLTTY] The brltty service gets in the way of other serial port applications
On 15/12/23 10:23, John wrote: I am new to this mailing list and recently had an experience with brltty that I would like to report. I can't see any access to historical posts so am unable to determine whether this subject has been mentioned already so I apologies upfront fpor any duplication. The topic has been mentioned before, and the BRLTTY developers are doing their best to keep BRLTTY working correctly without interfering with devices that are not braille displays. Whereas I appreciate the necessity of having such software for those who need Braille, I don't understand why it was installed automatically as a mainstream service as part of an OS update? I would expect such a program to be installed on an as-needs basis? For some of those who rely on BRLTTY, it is not possible to interact with the system at all without the braille support, so they can't install optional software without the competent help of another person. Such help may not be easily available, depending on the individual's circumstances, which is why the braille needs to be available by default, not as an optional extra. Secondly, disabling the service ought to have been sufficient to prevent the software from disrupting the operation of the system, but it wasn't. It had to be completely removed. Is this a bug? If so, where do I report it? I see the 'issues' function on the Github repository is not enabled. Perhaps your Linux distribution's bug tracker, if it's ultimately a packaging issue. Finally, if a program or update is likely to disrupt what might generally be considered 'normal' operation of the system (I "generally" because I accept that for users of Braille, having brltty installed is their 'normal'), shouldn't the software at least warn the user of this and request confirmation before installing itself? No, it shouldn't. If the braille isn't working, some users won't even know that such a prompt is there. They can't read anything that would be displayed on screen, so they can't read the prompt or know how to respond to it. I would respectfully suggest thinking this through from the perspective of someone who has no other means of using the operating system at all, if their braille display is not presenting the output. ___ This message was sent via the BRLTTY mailing list. To post a message, send an e-mail to: BRLTTY@brltty.app For general information, go to: http://brltty.app/mailman/listinfo/brltty
[BRLTTY] The brltty service gets in the way of other serial port applications
I am new to this mailing list and recently had an experience with brltty that I would like to report. I can't see any access to historical posts so am unable to determine whether this subject has been mentioned already so I apologies upfront fpor any duplication. I recently found that my I could not connect to my Arduino boards as they were no longer showing up as serial ports on my PC running Linux Mint. On further investigation I found the brltty and also brltty-udev service running on the machine. I did not have this running before and did not request this to be installed, but somehow there it was, presumably installed by an OS update. As it turns out, this is a known issue on the Arduino discussion forums. As per advice, I disabled both services, but this unfortunately did not restore normality. In fact, the service was still being loaded as could be evidenced by running the 'ps -ef' command to show a list of running processes. It runs out that the disabling the service was not sufficient and the brltty and brltty-udev packages needed to be un-installed from the system. Only then and finally after a re-boot, the boards started to show up as normal and I could connect to them. I should point out that Arduino is not the only area of electronics that I work with, that also requires access to serial ports. Whereas I appreciate the necessity of having such software for those who need Braille, I don't understand why it was installed automatically as a mainstream service as part of an OS update? I would expect such a program to be installed on an as-needs basis? Secondly, disabling the service ought to have been sufficient to prevent the software from disrupting the operation of the system, but it wasn't. It had to be completely removed. Is this a bug? If so, where do I report it? I see the 'issues' function on the Github repository is not enabled. Thirdly, can the software be configured to ignore certain ports? Finally, if a program or update is likely to disrupt what might generally be considered 'normal' operation of the system (I "generally" because I accept that for users of Braille, having brltty installed is their 'normal'), shouldn't the software at least warn the user of this and request confirmation before installing itself? Perhaps this has been done with the intention of making it as seamless as possible for Braille users to be able to make use of the program so I mean no disrespect, but I am curious as to how and why it suddenly appeared on my system. -- John. ___ This message was sent via the BRLTTY mailing list. To post a message, send an e-mail to: BRLTTY@brltty.app For general information, go to: http://brltty.app/mailman/listinfo/brltty
