Re: Orca does not speak
orca when the -r switch is used replaces its last process with a new process. On Tue, 15 Jan 2019, Linux for blind general discussion wrote: > Date: Tue, 15 Jan 2019 12:55:56 > From: Linux for blind general discussion > To: Linux for blind general discussion > Subject: Re: Orca does not speak > > OK, one more nit on this argument ... > > Linux for blind general discussion writes: > > Typing "orca -r", you kill this process (i.e., you remove it from > > the RAM), and you replace it with a new one. > > > The reason this is flawed is that there is no longer a Orca running once > the pid has been killed. Restarting Orca involves assigning a new pid to > it for inter-process communications. But, that's not a replacement, it's > an application restart that necessarily includes acquiring a process id. > > Now, if you could magically give Orca a new pid without killing the app, > then perhaps replace might be appropriate. > > ___ > Blinux-list mailing list > Blinux-list@redhat.com > https://www.redhat.com/mailman/listinfo/blinux-list > -- ___ Blinux-list mailing list Blinux-list@redhat.com https://www.redhat.com/mailman/listinfo/blinux-list
Re: Orca does not speak
I'll do my best to release it before my 70th birthday, Devin Didier On 15/01/2019 20:31, Linux for blind general discussion wrote: > Yes, I am eagerly awaiting said batch of updates, Didier. > > Devin Prater > >> On Jan 15, 2019, at 12:21 PM, Linux for blind general discussion >> wrote: >> >> I think that we disagree because we are not speaking of the same thing. >> >> When I write "replace orca" I mean: replace an instance of the orca >> application living in RAM. >> >> When you write "restart orca" you mean: restart orca, application >> usually stored on a mass storage device. >> >> To illustrate the difference: you can have two instances of orca >> running at the same time (one for Janina, one for Didier), but you >> have only one Orca installed in your system. >> >> This being said, I'd better spend my time preparing the next big >> batch of updates for Slit than arguing here, even more so as it's >> up to Joanie, not me, to decide of the option's name >> >> Best, >> >> Didier >> >> On 15/01/2019 18:55, Linux for blind general discussion wrote: >>> OK, one more nit on this argument ... >>> >>> Linux for blind general discussion writes: Typing "orca -r", you kill this process (i.e., you remove it from the RAM), and you replace it with a new one. >>> The reason this is flawed is that there is no longer a Orca running once >>> the pid has been killed. Restarting Orca involves assigning a new pid to >>> it for inter-process communications. But, that's not a replacement, it's >>> an application restart that necessarily includes acquiring a process id. >>> >>> Now, if you could magically give Orca a new pid without killing the app, >>> then perhaps replace might be appropriate. >>> >>> ___ >>> Blinux-list mailing list >>> Blinux-list@redhat.com >>> https://www.redhat.com/mailman/listinfo/blinux-list >>> >> >> ___ >> Blinux-list mailing list >> Blinux-list@redhat.com >> https://www.redhat.com/mailman/listinfo/blinux-list > > > ___ > Blinux-list mailing list > Blinux-list@redhat.com > https://www.redhat.com/mailman/listinfo/blinux-list > ___ Blinux-list mailing list Blinux-list@redhat.com https://www.redhat.com/mailman/listinfo/blinux-list
Re: Orca does not speak
Yes, I am eagerly awaiting said batch of updates, Didier. Devin Prater > On Jan 15, 2019, at 12:21 PM, Linux for blind general discussion > wrote: > > I think that we disagree because we are not speaking of the same thing. > > When I write "replace orca" I mean: replace an instance of the orca > application living in RAM. > > When you write "restart orca" you mean: restart orca, application > usually stored on a mass storage device. > > To illustrate the difference: you can have two instances of orca > running at the same time (one for Janina, one for Didier), but you > have only one Orca installed in your system. > > This being said, I'd better spend my time preparing the next big > batch of updates for Slit than arguing here, even more so as it's > up to Joanie, not me, to decide of the option's name > > Best, > > Didier > > On 15/01/2019 18:55, Linux for blind general discussion wrote: >> OK, one more nit on this argument ... >> >> Linux for blind general discussion writes: >>> Typing "orca -r", you kill this process (i.e., you remove it from >>> the RAM), and you replace it with a new one. >>> >> The reason this is flawed is that there is no longer a Orca running once >> the pid has been killed. Restarting Orca involves assigning a new pid to >> it for inter-process communications. But, that's not a replacement, it's >> an application restart that necessarily includes acquiring a process id. >> >> Now, if you could magically give Orca a new pid without killing the app, >> then perhaps replace might be appropriate. >> >> ___ >> Blinux-list mailing list >> Blinux-list@redhat.com >> https://www.redhat.com/mailman/listinfo/blinux-list >> > > ___ > Blinux-list mailing list > Blinux-list@redhat.com > https://www.redhat.com/mailman/listinfo/blinux-list ___ Blinux-list mailing list Blinux-list@redhat.com https://www.redhat.com/mailman/listinfo/blinux-list
Re: Orca does not speak
I think that we disagree because we are not speaking of the same thing. When I write "replace orca" I mean: replace an instance of the orca application living in RAM. When you write "restart orca" you mean: restart orca, application usually stored on a mass storage device. To illustrate the difference: you can have two instances of orca running at the same time (one for Janina, one for Didier), but you have only one Orca installed in your system. This being said, I'd better spend my time preparing the next big batch of updates for Slit than arguing here, even more so as it's up to Joanie, not me, to decide of the option's name Best, Didier On 15/01/2019 18:55, Linux for blind general discussion wrote: > OK, one more nit on this argument ... > > Linux for blind general discussion writes: >> Typing "orca -r", you kill this process (i.e., you remove it from >> the RAM), and you replace it with a new one. >> > The reason this is flawed is that there is no longer a Orca running once > the pid has been killed. Restarting Orca involves assigning a new pid to > it for inter-process communications. But, that's not a replacement, it's > an application restart that necessarily includes acquiring a process id. > > Now, if you could magically give Orca a new pid without killing the app, > then perhaps replace might be appropriate. > > ___ > Blinux-list mailing list > Blinux-list@redhat.com > https://www.redhat.com/mailman/listinfo/blinux-list > ___ Blinux-list mailing list Blinux-list@redhat.com https://www.redhat.com/mailman/listinfo/blinux-list
Re: Orca does not speak
I'd guess that in speech most people would say re start Orca with the command "orca -r" or "orca --replace", and everyone would be and has been OK with that. This only came up when someone, who probably never read the man page for Orca, said that using the word "replace" for a command line switch was using English improperly and was imprecise. It seems to me that we're now talking details about what is technically meant by the words "replace" and "restart" in computer science. I think we've moved past whether we're using the English language correctly or not and whether we're being precise in a layman's view. IMHO, the Orca developer has bigger fish on her plate then whether a command line flag is properly using English or not, but if anyone feels strongly enough about this they should bring it up on the Orca list and have the command line flag and associated documentation updated. I may not use the word "replace" in this context myself, but I'm not going to quibble with another developer who chooses to use the word "replace" in this context. I think someone also mentioned how surprised they were that Orca has command line flags. I just assume everything in Linux has command line flags, and I'd be surprised to stumble across a command or an application that didn't have a set of command line flags. Just try typing your favorite command or application followed by the "--help" flag and see what you get. For example, try "firefox --help" or "google-chrome --help" or "rhythmbox --help". For completeness, here's what the Orca man page says about the replace flag: *--replace* Replace a currently running*orca* process. By default, if*orca* detects an existing *orca* process for the same session, it will not start a new*orca* process. This option will kill and cleanup after any existing*orca* process and then start a new *orca* in its place. On 1/15/19 7:27 AM, Linux for blind general discussion wrote: Maybe there's some subtle distinction I'm not catching, but saying it kills the running process and replaces it with a new one sounds like a convoluted way of saying it restarts the process, and I think most who aren't trying to justify the wording of the switch would say it restarts the program. Actually, if I remember correctly, the original answer to what the --replace switch does was "it restarts orca" or something to that effect, and the more detailed answer only came up when someone pointed out the odd wording. --replace might be technically correct, but it still strikes me as using a word in an unusual context most won't understand without explanation when a different word would get the meaning across without explanation. Kind of reminds me of how Americans sometimes have trouble understanding Brits because of common words that vary greatly in their common definition on opposite sides of the pond(and for all I know, replace might be commonly understood in this context in some part of the anglosphere other than my own). I understand the explanation for why the switch is --replace, but I'd probably still call it --restart if I was going to include such functionality in a program I wrote myself. On a more humorous note, without the context that orca -r restarts orca, I'd probably be wondering what a screen reader could possibly reverse or recurse since those are the most common things a -r or -R switch do. --Jeff ___ Blinux-list mailing list Blinux-list@redhat.com https://www.redhat.com/mailman/listinfo/blinux-list -- Christopher (CJ) Chaltain at Gmail ___ Blinux-list mailing list Blinux-list@redhat.com https://www.redhat.com/mailman/listinfo/blinux-list
Re: Orca does not speak
OK, one more nit on this argument ... Linux for blind general discussion writes: > Typing "orca -r", you kill this process (i.e., you remove it from > the RAM), and you replace it with a new one. > The reason this is flawed is that there is no longer a Orca running once the pid has been killed. Restarting Orca involves assigning a new pid to it for inter-process communications. But, that's not a replacement, it's an application restart that necessarily includes acquiring a process id. Now, if you could magically give Orca a new pid without killing the app, then perhaps replace might be appropriate. ___ Blinux-list mailing list Blinux-list@redhat.com https://www.redhat.com/mailman/listinfo/blinux-list
Re: Orca does not speak
Hmmm, a college level course in English might serve better. Linux for blind general discussion writes: > There are two subtle distinctions you're not catching. > First when a process gets killed and then replaced that process gives up > its original process id number and when the process is replaced, it gets > a new usually higher process id number. Sometimes another process may > conflict with the lower process id number and screen-reader/orca getting > a higher number escapes that conflict. So replace is absolutely > technically correct but you have to learn more about computer hardware > and computer software to understand that. A college class on computer > hardware/software can help in this regard. > > On Tue, 15 Jan 2019, Linux for blind general discussion wrote: > > > Date: Tue, 15 Jan 2019 08:27:03 > > From: Linux for blind general discussion > > To: blinux-list@redhat.com > > Subject: Re: Orca does not speak > > > > Maybe there's some subtle distinction I'm not catching, but saying it > > kills the running process and replaces it with a new one sounds like a > > convoluted way of saying it restarts the process, and I think most who > > aren't trying to justify the wording of the switch would say it > > restarts the program. Actually, if I remember correctly, the original > > answer to what the --replace switch does was "it restarts orca" or > > something to that effect, and the more detailed answer only came up > > when someone pointed out the odd wording. > > > > --replace might be technically correct, but it still strikes me as > > using a word in an unusual context most won't understand without > > explanation when a different word would get the meaning across without > > explanation. Kind of reminds me of how Americans sometimes have > > trouble understanding Brits because of common words that vary greatly > > in their common definition on opposite sides of the pond(and for all I > > know, replace might be commonly understood in this context in some > > part of the anglosphere other than my own). > > > > I understand the explanation for why the switch is --replace, but I'd > > probably still call it --restart if I was going to include such > > functionality in a program I wrote myself. > > > > On a more humorous note, without the context that orca -r restarts > > orca, I'd probably be wondering what a screen reader could possibly > > reverse or recurse since those are the most common things a -r or -R > > switch do. > > > > --Jeff > > > > ___ > > Blinux-list mailing list > > Blinux-list@redhat.com > > https://www.redhat.com/mailman/listinfo/blinux-list > > > > -- > > ___ > Blinux-list mailing list > Blinux-list@redhat.com > https://www.redhat.com/mailman/listinfo/blinux-list -- Janina Sajka Linux Foundation Fellow Executive Chair, Accessibility Workgroup: http://a11y.org The World Wide Web Consortium (W3C), Web Accessibility Initiative (WAI) Chair, Accessible Platform Architectureshttp://www.w3.org/wai/apa ___ Blinux-list mailing list Blinux-list@redhat.com https://www.redhat.com/mailman/listinfo/blinux-list
Re: Orca does not speak
Well, and let me note where we do commonly use the term replace in computing. If I've been using mutt to read email and then suddently start using Thunderbird instead, we would say I replaced my email application. If I've been using bash and decided to switch to zsh, we would say I replaced bash with zsh, I replaced my shell. Janina Linux for blind general discussion writes: > Maybe there's some subtle distinction I'm not catching, but saying it > kills the running process and replaces it with a new one sounds like a > convoluted way of saying it restarts the process, and I think most who > aren't trying to justify the wording of the switch would say it > restarts the program. Actually, if I remember correctly, the original > answer to what the --replace switch does was "it restarts orca" or > something to that effect, and the more detailed answer only came up > when someone pointed out the odd wording. > > --replace might be technically correct, but it still strikes me as > using a word in an unusual context most won't understand without > explanation when a different word would get the meaning across without > explanation. Kind of reminds me of how Americans sometimes have > trouble understanding Brits because of common words that vary greatly > in their common definition on opposite sides of the pond(and for all I > know, replace might be commonly understood in this context in some > part of the anglosphere other than my own). > > I understand the explanation for why the switch is --replace, but I'd > probably still call it --restart if I was going to include such > functionality in a program I wrote myself. > > On a more humorous note, without the context that orca -r restarts > orca, I'd probably be wondering what a screen reader could possibly > reverse or recurse since those are the most common things a -r or -R > switch do. > > --Jeff > > ___ > Blinux-list mailing list > Blinux-list@redhat.com > https://www.redhat.com/mailman/listinfo/blinux-list -- Janina Sajka Linux Foundation Fellow Executive Chair, Accessibility Workgroup: http://a11y.org The World Wide Web Consortium (W3C), Web Accessibility Initiative (WAI) Chair, Accessible Platform Architectureshttp://www.w3.org/wai/apa ___ Blinux-list mailing list Blinux-list@redhat.com https://www.redhat.com/mailman/listinfo/blinux-list
Re: Orca does not speak
This is not convincing, Didier. Note my example of Control Alt Delete. We always call that restart. We never call it replace. It also gives us an entire new set of pids, even for the same apps. If I'm reading some text file with less and kill it with Control C, I can restart the same app reading the same file by pressing up arrow and enter. We always call that a restart. Apps are restarted in standard English usage. They're not replaced. Janina Linux for blind general discussion writes: > Hello Janina, > > technically your screen reader is a process in RAM, communicating > with other processes, like at-spi and speech-dispatcher. > > Typing "orca -r", you kill this process (i.e., you remove it from > the RAM), and you replace it with a new one. > > Here is an example. > > In the a terminal I type this command: > > didier[~]$ ps -C orca > PID TTY TIME CMD > 26823 tty1 00:00:34 orca > > In the output, PID is the process identifier. > > Then, I press Alt+F2 and type orca -r > > After that: > PID TTY TIME CMD > 394 tty1 00:00:00 orca > didier[~]$ > > > So the previous process disappeared and we have a new one. > > So technically, it is indeed a replacement, not a restart: > I replaced a screen reader by a new one. > > Best, > > Didier > > > On 15/01/2019 08:02, Linux for blind general discussion wrote: > > I rely on "orca -r" quite a bit, actually. There are any number of > > conditions that can silence a running Orca that can be quickly remedied > > that way. Guess I never noticed I stood for "replace," and the > > juxtaposition of "screen-reader" really threw me. Replace my screen > > reader? With what? > > > > Best, > > > > Janina > > > > Linux for blind general discussion writes: > >> For what it's worth, I'd agree --restart would be more intuitive and > >> self-documenting from the perspective of a native English speaker > >> assuming there isn't already a --restart switch that does something > >> different. That said, it's ultimately the developer's decision, it's > >> hardly the most esoteric command-line switch in existence, and > >> honestly, I'm kind of surprised a graphical application would even > >> have such a function built-in to its command-line syntax. > >> > >> ___ > >> Blinux-list mailing list > >> Blinux-list@redhat.com > >> https://www.redhat.com/mailman/listinfo/blinux-list > > > > ___ > Blinux-list mailing list > Blinux-list@redhat.com > https://www.redhat.com/mailman/listinfo/blinux-list -- Janina Sajka Linux Foundation Fellow Executive Chair, Accessibility Workgroup: http://a11y.org The World Wide Web Consortium (W3C), Web Accessibility Initiative (WAI) Chair, Accessible Platform Architectureshttp://www.w3.org/wai/apa ___ Blinux-list mailing list Blinux-list@redhat.com https://www.redhat.com/mailman/listinfo/blinux-list
Re: Orca does not speak
System admin classes if done for linux or unix cover process ids along with some system utilities to control them when that becomes necessary. Whether or not you've had a system admin class in Linux, if you've installed and have your own Linux system operating congratulations, you just got a Linux system Administrator's Hat. Most of those virtual hats have amateur status on them and some novice and a few have professional status. On Tue, 15 Jan 2019, Linux for blind general discussion wrote: > Date: Tue, 15 Jan 2019 11:48:57 > From: Linux for blind general discussion > To: blinux-list@redhat.com > Subject: Re: Orca does not speak > > I actually hold a BS in Computer Science, though I confess none of my > classes ever went into much detail regarding process IDs, not even the > ones that dealt with Linux(granted, even the Linux-heavy classes > seemed designed for students coming from a Windows background, and by > the time I finished my general education studies and started focusing > on my major, I had been a full-time Linux user for a few years. > > That said, while I can understand the reasoning, and might could even > agree it's the most reasonable choice from an internal, developer > facing perspective, I've been mostly approaching this discussion from > an external, end-user perspective, the explanation requires delving > into what should arguably be a black box from the end-user's side, and > if the wording confuses people who are likely well-above average in > being technologically literate, I feel bad for those who are learning > to use Linux or a computer for the first time. > > I won't begrudge a developer, not even a FOSS developer, for > optimizing their code for their own readibility and workflow, > especially when they expect few external contributions even if they > optimize for general readability, but I do think UI should strive for > minimizing the amount of technical knowledge the end user needs in > order to understand how to operate the software, especially for > something with as general use as a screen reader. > > But meh, I feel like both sides are repeating themselves, neither is > likely to change their opinion, and it's not likely I or anyone else > will make a personal fork of orca to change a pair of two-character > substrings in a single long switch. Wasn't even aware Orca had a > command-line interface prior to this thread and I usually just fix > issues with orca by either closing Firefox, which given my setup > triggers termination of orca and my x-session, or if that fails, > rebooting the computer, which is usually sufficient to fix minor > problems or point out problems big enough to restore a backup of my > root partition(though admittedly, that latter definitely isn't > something I'd expect a beginner to be able to do and I manage such > through bash scripts that save me from having to memorize the cryptic > syntax of partimage). > > ___ > Blinux-list mailing list > Blinux-list@redhat.com > https://www.redhat.com/mailman/listinfo/blinux-list > -- ___ Blinux-list mailing list Blinux-list@redhat.com https://www.redhat.com/mailman/listinfo/blinux-list
getslint.sh script
I wrote a script to pull the latest slint version and check the slint iso for integrity. I've verified the script works too. For this script to work it needs a bsd tag file with slint's valid sha1sum value in it so integrity checking is possible and that needs to be on the hard drive first. I'm sharing this script for two reasons. First, I solved some problems I didn't think were going to be possible to solve. Second, changing the slint iso download item in the script and the name of the sha1sum file so it matches another on a hard drive the script can be used to download other files from sourceforge.net and check their integrity too. If a bad download happens, the script deletes the bad iso and continues to try to download a good iso and you get information when the download starts and when it ends. There is also a fail count which is always printed out if any download fails integrity check, you find out how many times that happened. -- cut here. #!/usr/bin/env bash # file: getslint.sh - download latest slint iso. # set fail count. FC=0 # remove old wget-log file if present. if [ -f wget-log ]; then rm wget-log fi # set integrity check variable to error value of 1. IC=1 # enter integrity check loop. until [ $IC -lt 1 ]; do # show starting time stamp. echo "starting download at:" date -R # download latest slint iso. wget -bc --tries=inf --max-redirect=1 --trust-server-names https://sourceforge.net/projects/slint/files/latest/download # set DL variable to error value 1. DL=1 until [ $DL -lt 1 ]; do # allow some time for download. sleep 15 # search file wget-log for saved string. grep "saved" wget-log # check grep return value for zero value meaning saved line was found. if [ $? -lt 1 ]; then # clear error value from DL switch. let DL-=1 fi done # do integrity check. sha256sum -c slint64-14.2.1.1.sha256sum # if return of sha1sum is zero, remove error value from integrity check switch. if [ $? -lt 1 ]; then let IC-=1 echo -n "fail count: " echo $FC else let FC+=1 rm slint*.iso rm wget-log fi done cut here. ___ Blinux-list mailing list Blinux-list@redhat.com https://www.redhat.com/mailman/listinfo/blinux-list
Re: Orca does not speak
I actually hold a BS in Computer Science, though I confess none of my classes ever went into much detail regarding process IDs, not even the ones that dealt with Linux(granted, even the Linux-heavy classes seemed designed for students coming from a Windows background, and by the time I finished my general education studies and started focusing on my major, I had been a full-time Linux user for a few years. That said, while I can understand the reasoning, and might could even agree it's the most reasonable choice from an internal, developer facing perspective, I've been mostly approaching this discussion from an external, end-user perspective, the explanation requires delving into what should arguably be a black box from the end-user's side, and if the wording confuses people who are likely well-above average in being technologically literate, I feel bad for those who are learning to use Linux or a computer for the first time. I won't begrudge a developer, not even a FOSS developer, for optimizing their code for their own readibility and workflow, especially when they expect few external contributions even if they optimize for general readability, but I do think UI should strive for minimizing the amount of technical knowledge the end user needs in order to understand how to operate the software, especially for something with as general use as a screen reader. But meh, I feel like both sides are repeating themselves, neither is likely to change their opinion, and it's not likely I or anyone else will make a personal fork of orca to change a pair of two-character substrings in a single long switch. Wasn't even aware Orca had a command-line interface prior to this thread and I usually just fix issues with orca by either closing Firefox, which given my setup triggers termination of orca and my x-session, or if that fails, rebooting the computer, which is usually sufficient to fix minor problems or point out problems big enough to restore a backup of my root partition(though admittedly, that latter definitely isn't something I'd expect a beginner to be able to do and I manage such through bash scripts that save me from having to memorize the cryptic syntax of partimage). ___ Blinux-list mailing list Blinux-list@redhat.com https://www.redhat.com/mailman/listinfo/blinux-list
Re: Orca does not speak
There are two subtle distinctions you're not catching. First when a process gets killed and then replaced that process gives up its original process id number and when the process is replaced, it gets a new usually higher process id number. Sometimes another process may conflict with the lower process id number and screen-reader/orca getting a higher number escapes that conflict. So replace is absolutely technically correct but you have to learn more about computer hardware and computer software to understand that. A college class on computer hardware/software can help in this regard. On Tue, 15 Jan 2019, Linux for blind general discussion wrote: > Date: Tue, 15 Jan 2019 08:27:03 > From: Linux for blind general discussion > To: blinux-list@redhat.com > Subject: Re: Orca does not speak > > Maybe there's some subtle distinction I'm not catching, but saying it > kills the running process and replaces it with a new one sounds like a > convoluted way of saying it restarts the process, and I think most who > aren't trying to justify the wording of the switch would say it > restarts the program. Actually, if I remember correctly, the original > answer to what the --replace switch does was "it restarts orca" or > something to that effect, and the more detailed answer only came up > when someone pointed out the odd wording. > > --replace might be technically correct, but it still strikes me as > using a word in an unusual context most won't understand without > explanation when a different word would get the meaning across without > explanation. Kind of reminds me of how Americans sometimes have > trouble understanding Brits because of common words that vary greatly > in their common definition on opposite sides of the pond(and for all I > know, replace might be commonly understood in this context in some > part of the anglosphere other than my own). > > I understand the explanation for why the switch is --replace, but I'd > probably still call it --restart if I was going to include such > functionality in a program I wrote myself. > > On a more humorous note, without the context that orca -r restarts > orca, I'd probably be wondering what a screen reader could possibly > reverse or recurse since those are the most common things a -r or -R > switch do. > > --Jeff > > ___ > Blinux-list mailing list > Blinux-list@redhat.com > https://www.redhat.com/mailman/listinfo/blinux-list > -- ___ Blinux-list mailing list Blinux-list@redhat.com https://www.redhat.com/mailman/listinfo/blinux-list
Re: Orca does not speak
Maybe there's some subtle distinction I'm not catching, but saying it kills the running process and replaces it with a new one sounds like a convoluted way of saying it restarts the process, and I think most who aren't trying to justify the wording of the switch would say it restarts the program. Actually, if I remember correctly, the original answer to what the --replace switch does was "it restarts orca" or something to that effect, and the more detailed answer only came up when someone pointed out the odd wording. --replace might be technically correct, but it still strikes me as using a word in an unusual context most won't understand without explanation when a different word would get the meaning across without explanation. Kind of reminds me of how Americans sometimes have trouble understanding Brits because of common words that vary greatly in their common definition on opposite sides of the pond(and for all I know, replace might be commonly understood in this context in some part of the anglosphere other than my own). I understand the explanation for why the switch is --replace, but I'd probably still call it --restart if I was going to include such functionality in a program I wrote myself. On a more humorous note, without the context that orca -r restarts orca, I'd probably be wondering what a screen reader could possibly reverse or recurse since those are the most common things a -r or -R switch do. --Jeff ___ Blinux-list mailing list Blinux-list@redhat.com https://www.redhat.com/mailman/listinfo/blinux-list
Re: Orca does not speak
Hello Janina, technically your screen reader is a process in RAM, communicating with other processes, like at-spi and speech-dispatcher. Typing "orca -r", you kill this process (i.e., you remove it from the RAM), and you replace it with a new one. Here is an example. In the a terminal I type this command: didier[~]$ ps -C orca PID TTY TIME CMD 26823 tty1 00:00:34 orca In the output, PID is the process identifier. Then, I press Alt+F2 and type orca -r After that: PID TTY TIME CMD 394 tty1 00:00:00 orca didier[~]$ So the previous process disappeared and we have a new one. So technically, it is indeed a replacement, not a restart: I replaced a screen reader by a new one. Best, Didier On 15/01/2019 08:02, Linux for blind general discussion wrote: > I rely on "orca -r" quite a bit, actually. There are any number of > conditions that can silence a running Orca that can be quickly remedied > that way. Guess I never noticed I stood for "replace," and the > juxtaposition of "screen-reader" really threw me. Replace my screen > reader? With what? > > Best, > > Janina > > Linux for blind general discussion writes: >> For what it's worth, I'd agree --restart would be more intuitive and >> self-documenting from the perspective of a native English speaker >> assuming there isn't already a --restart switch that does something >> different. That said, it's ultimately the developer's decision, it's >> hardly the most esoteric command-line switch in existence, and >> honestly, I'm kind of surprised a graphical application would even >> have such a function built-in to its command-line syntax. >> >> ___ >> Blinux-list mailing list >> Blinux-list@redhat.com >> https://www.redhat.com/mailman/listinfo/blinux-list > ___ Blinux-list mailing list Blinux-list@redhat.com https://www.redhat.com/mailman/listinfo/blinux-list
