A few issues I have with recent changes.

[AConcernedThirdParty]

Not criticising the changes made - In fact I think they're brilliant ideas, but I have problems with a few things.

Firstly, in speech.js, I have a feeling that a few phrases may not be in tune with the theme of the show, (i.e. phrases Samaritan would say). For example, "Say my name", "Heisenberg".

Secondly, I believe the user should get a choice as to whether speech recognition is activated on each page load. Granted, the user is asked whether to grant Samaritan access to their microphone on first load (by the browser), but it's not asked afterwards. Perhaps only enabling it through the "speech" url parameter?

Thirdly, a few formatting/file location problems.

• The root directory is a bit cluttered with CSS and JS. It may be a good idea to organise these into folders. This wouldn't really affect how the site works, but makes it more presentable to anyone editing it.
• The code in speech.js (particularly lines 13-28) is a bit too cluttered. It makes it hard to see which phrases are available, and the text displayed when each phrase is said. Again, it may not affect the site at all, but it makes the file easier to edit.

Finally, it'd be nice if a .woff, in addition to the .ttf is added. This would allow the MagdaCleanMono font to show up on Firefox/IE/Other browsers not supporting .ttf.

I would write some code and tackle a few of these issues, but I'm currently out of the country and typing this up on my mobile.

[rodrigograca31]

I agree with everything you mentioned.....
@AlexandruSimandi What do you think about the speech things?

I'm at work, but tomorrow I can tackle these "problems".

The URL parameter is a good idea, but it's "for us" that know how to do it... it's a bit hard for "Normal" users but I think I will add that and the "msg" parameter to the README file...

[AlexandruSimandi]

@AConcernedThirdParty Thanks for the code feedback. I was on a coding spree that night and i totally agree it could be done better.

Firstly, in speech.js, I have a feeling that a few phrases may not be in tune with the theme of the show, (i.e. phrases Samaritan would say). For example, "Say my name", "Heisenberg".

I did that for fun, will remove it. From now on only show related content.

Secondly, I believe the user should get a choice as to whether speech recognition is activated on each page load. Granted, the user is asked whether to grant Samaritan access to their microphone on first load (by the browser), but it's not asked afterwards. Perhaps only enabling it through the "speech" url parameter?

Chrome used to ask permission at every load. I find it more immersive not to be asked every time, it's distracting from the whole Samaritan vibe. Speech url parameter could totally be done but how should a regular POI reddit fan know how to add or remove param?
Also, it's running on the client side. Someone should hack Chrome to eavesdrop.

The code in speech.js (particularly lines 13-28) is a bit too cluttered. It makes it hard to see which phrases are available, and the text displayed when each phrase is said. Again, it may not affect the site at all, but it makes the file easier to edit.

The only other way is too make a separate json file and add them dynamically. This, I will totally do.

[rodrigograca31]

I've tackled most of these issues.

I've not done the "microphone URL parameter" thing...

I've found something cool: http://samaritian.herokuapp.com/

And I'm done for today... :)

[AdminOdd]

Question Bro. If you know about the problems existing. Why don't you take
the time to fix it and you know make it better :D No critium the about the
bugs found - In fact I think they're good you found them, but U could fix
it you know :D

On Sat, Jun 25, 2016 at 2:40 PM, Rodrigo Graça [email protected]
wrote:

I've tackled most of these issues.

I've not done the "microphone URL parameter" thing...

I've found something cool: http://samaritian.herokuapp.com/

And I'm done for today... :)

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
#20 (comment),
or mute the thread
https://github.com/notifications/unsubscribe/ASemlUmiYArhj416Ep3_9jUaOhg6YWILks5qPXYagaJpZM4I9suR
.

[AdminOdd]

Question Bro. If you know about the problems existing. Why don't you take
the time to fix it and you know make it better :D No critium the about the
bugs found - In fact I think they're good you found them, but U could fix
it you know :D

On Fri, Jun 24, 2016 at 7:50 AM, AConcernedThirdParty <
[email protected]> wrote:

Not criticising the changes made - In fact I think they're brilliant
ideas, but I have problems with a few things.

Firstly, in speech.js
https://github.com/rodrigograca31/Samaritan/blob/master/speech.js, I
have a feeling that a few phrases may not be in tune with the theme of the
show, (i.e. phrases Samaritan would say). For example, "Say my name",
"Heisenberg".

Secondly, I believe the user should get a choice as to whether speech
recognition is activated on each page load. Granted, the user is asked
whether to grant Samaritan access to their microphone on first load (by the
browser), but it's not asked afterwards. Perhaps only enabling it through
the "speech" url parameter?

Thirdly, a few formatting/file location problems.

• The root directory is a bit cluttered with CSS and JS. It may be a
  good idea to organise these into folders. This wouldn't really affect how
  the site works, but makes it more presentable to anyone editing it.
• The code in speech.js (particularly lines 13-28) is a bit too
  cluttered. It makes it hard to see which phrases are available, and the
  text displayed when each phrase is said. Again, it may not affect the site
  at all, but it makes the file easier to edit.

Finally, it'd be nice if a .woff, in addition to the .ttf is added. This
would allow the MagdaCleanMono font to show up on Firefox/IE/Other browsers
not supporting .ttf.

I would write some code and tackle a few of these issues, but I'm
currently out of the country and typing this up on my mobile.

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
#20, or mute the
thread
https://github.com/notifications/unsubscribe/ASemlRxmuamjNp7DefEWlwE1SHr_hEN4ks5qO8SFgaJpZM4I9suR
.

[AlexandruSimandi]

He just said he is out of country