Charles McNulty wrote:
Well, obviously RC is highly dependent on javascript, which is totally browser dependent. I don't know to what degree if any mobile browsers implement javascript. In other words, single-clicking vs. double-clicking on a cell phone is just the tip of the iceberg in terms of problems using RC on a cell phone. I guess the question is, what browser does the cell phone use? (if it's even possible to get that info) , does that browser implement javascript? and is there a way to run that browser on a workstation instead of just a cell so that people can test it without having to get the phone?
-Charles
From my web logs, my phone reports itself as:
SAMSUNG-SGH-D500/1.0 Profile/MIDP-2.0 Configuration/CLDC-1.1 UP.Browser/6.2.3.3.c.1.101 (GUI) MMP/2.0
But no, it doesn't support javascript.
I guess that you don't subscribe to the belief of gracious degradation then. To me javascript, while being capable of performing wonderful usability enhancing functions, should never be a requirement. Web pages and applications should be able to function perfectly well with JS disabled; JS can certainly be used to make things easier and flashier, but not at the cost of accessibility. I thoroughly recommend reading A List Apart http://www.alistapart.com; they have many many articles about accessibility, none of which advocates reducing your web site / application to an archaic state all in the name of accessibility.
I realise that the majority of the functions within the application rely on javascript and I am certainly not trying to say it should all be rewritten, but I think that you should at least provide the ability to READ e-mails for those people using browsers without javascript support.
Kevin L. wrote:
I totally understand and agree with Charles' response, and the last thing I would like to see would be hyper links all over the place as the current setup is quite beautiful
Never heard of CSS? How can something as innocuous as a simple hyperlink make a page ugly? If it's the underlines that bother you, CSS has a remarkable ability that can come to the rescue here...
I will have a play with this this weekend and I will try and see how the application functions with added hyperlinks. If I think I can get it to work without breaking any other functions then I will submit the changes to this list for comment.
Thanks, Richard
Richard Green wrote:
I guess that you don't subscribe to the belief of gracious degradation then. To me javascript, while being capable of performing wonderful usability enhancing functions, should never be a requirement. Web pages and applications should be able to function perfectly well with JS disabled; JS can certainly be used to make things easier and flashier, but not at the cost of accessibility. I thoroughly recommend reading A
Well, I draw a distinction between an application and a page. A page's
primary function is to distribute information. It should absolutely be
accessible to the widest possible audience and should never require
javascript in order to view the information it's attempting to present.
Heck until a couple years ago I still tested my web pages in lynx.
Applications on the other hand have always come with dependancies and
minimum requirements. Desktop applications are built for a specific
platform. They often require certain libraries (.net, java, gtk, etc).
It is in this vein that I see absolutely no problem with setting
javascript as a minimum requirement for roundcube.
The thing is that there are already a ton of webmail applications out there that don't rely on or even use javascript that would be ideal for a javascriptless environment. Roundcube doesn't just use javascript to make things pretty, javascript is it's entire raison d'etre. It's what sets it apart from the myriad of other webmail applications. You might argue that gmail is able to degrade somewhat gracefully without javascript, but keep in mind that gmail's primary purpose isn't to deliver e-mail. It's primary purpose is to deliver advertisement and generate revenue, therfore it must be accessible to the most people and it makes sense that they would require it to degrade well. Roundcube isn't in the business of delievering advertisement (thank goodness), and is therefore not constrained by the same need to degrade.
By the way, these are all just my opinions. I'm not in the driver's seat, and I haven't even developed that much code. I'm just opinionated.
-Charles
Richard Green wrote:
Charles McNulty wrote:
Well, obviously RC is highly dependent on javascript, which is totally browser dependent. I don't know to what degree if any mobile browsers implement javascript. In other words, single-clicking vs. double-clicking on a cell phone is just the tip of the iceberg in terms of problems using RC on a cell phone. I guess the question is, what browser does the cell phone use? (if it's even possible to get that info) , does that browser implement javascript? and is there a way to run that browser on a workstation instead of just a cell so that people can test it without having to get the phone?
-Charles
From my web logs, my phone reports itself as:
SAMSUNG-SGH-D500/1.0 Profile/MIDP-2.0 Configuration/CLDC-1.1 UP.Browser/6.2.3.3.c.1.101 (GUI) MMP/2.0
But no, it doesn't support javascript.
I guess that you don't subscribe to the belief of gracious degradation then. To me javascript, while being capable of performing wonderful usability enhancing functions, should never be a requirement. Web pages and applications should be able to function perfectly well with JS disabled; JS can certainly be used to make things easier and flashier, but not at the cost of accessibility. I thoroughly recommend reading A List Apart http://www.alistapart.com; they have many many articles about accessibility, none of which advocates reducing your web site / application to an archaic state all in the name of accessibility.
I realise that the majority of the functions within the application rely on javascript and I am certainly not trying to say it should all be rewritten, but I think that you should at least provide the ability to READ e-mails for those people using browsers without javascript support.
Kevin L. wrote:
I totally understand and agree with Charles' response, and the last thing I would like to see would be hyper links all over the place as the current setup is quite beautiful
Never heard of CSS? How can something as innocuous as a simple hyperlink make a page ugly? If it's the underlines that bother you, CSS has a remarkable ability that can come to the rescue here...
I will have a play with this this weekend and I will try and see how the application functions with added hyperlinks. If I think I can get it to work without breaking any other functions then I will submit the changes to this list for comment.
Thanks, Richard
Obviously, you never read the email from Charles McNulty about degredation of the JS. I agree with Charles in his sentiments that this is is a Web 2.0 project, and it's a desktop type application, not a web page. You can use SquirrelMail or Horde if you want. That doesn't use JS. But why should a project that on the front page has a requirement as JS degrade for those that can't handle it. Not only would it delay the final release date, but essentially you'd have another webmail client, and not the one that I think Thomas has envisioned.
Degrading the JS is a good idea for regular web pages, but this is a web application, not a page. This is meant to be interactive and use Web 2.0 technology. I'm sorry, but I don't see why we should stop down to lesser tech just to satisfy a phone.... most of our users will be using computers and can enable JS for their email site.
Brett Patters - Roundcube Forum Admin wrote:
Obviously, you never read the email from Charles McNulty about degredation of the JS. I agree with Charles in his sentiments that this is is a Web 2.0 project, and it's a desktop type application, not a web page. You can use SquirrelMail or Horde if you want. That doesn't use JS. But why should a project that on the front page has a requirement as JS degrade for those that can't handle it. Not only would it delay the final release date, but essentially you'd have another webmail client, and not the one that I think Thomas has envisioned.
Degrading the JS is a good idea for regular web pages, but this is a web application, not a page. This is meant to be interactive and use Web 2.0 technology. I'm sorry, but I don't see why we should stop down to lesser tech just to satisfy a phone.... most of our users will be using computers and can enable JS for their email site.
Ok, I have read all your arguments about the Web 2.0 application style thing and I agree with them.
However, I think it is very short-sighted not to consider accessibility issues. If I were unable to use a mouse due to handicap, I would not be able to use this application fully. All other windows applications are perfectly navigable using only a keyboard.
I know a lot of work has been done around keyboard navigation and I praise that; I am simply pointing out that one obvious omission is that you can't open an e-mail without double clicking on it. (At least I couldn't work out how to do it, but experience tells me that it should be possible using just tabs, arrow keys and pressing enter on the desired message; every other application works that way and so should this)
My suggestion of using hyperlinks seems the most obvious and easiest way of rectifying this problem and, as a plus, I would be able to check my e-mail on my mobile phone.
-- Richard Green
Richard Green wrote:
However, I think it is very short-sighted not to consider accessibility issues. If I were unable to use a mouse due to handicap, I would not be able to use this application fully. All other windows applications are perfectly navigable using only a keyboard.
I know a lot of work has been done around keyboard navigation and I praise that; I am simply pointing out that one obvious omission is that you can't open an e-mail without double clicking on it. (At least I couldn't work out how to do it, but experience tells me that it should be possible using just tabs, arrow keys and pressing enter on the desired message; every other application works that way and so should this)
My suggestion of using hyperlinks seems the most obvious and easiest way of rectifying this problem and, as a plus, I would be able to check my e-mail on my mobile phone.
I'm still against the hyperlinks, as I was the last time this got brought up months ago, but you do bring up an interesting point with keyboard accessability. I'm not running the latest CVS or anything, but would making the Enter button open whatever message is highlighted open be all that difficult?
-Trae Dorn
Trae Dorn wrote:
I'm still against the hyperlinks, as I was the last time this got brought up months ago, but you do bring up an interesting point with keyboard accessability. I'm not running the latest CVS or anything, but would making the Enter button open whatever message is highlighted open be all that difficult?
Actually I find that it's not possible to even _highlight_ a message without physically clicking it, so unfortunately there's quite a bit of work needed here.
I do understand peoples reluctance to the hyperlink proposition, but it really does seem to me the simplest way of overcoming the accessibility issue.
I'll give it some more thought; I'm certainly not about to ditch this app just because I can't use it on my mobile!
Richard Green wrote:
Brett Patters - Roundcube Forum Admin wrote:
Obviously, you never read the email from Charles McNulty about degredation of the JS. I agree with Charles in his sentiments that this is is a Web 2.0 project, and it's a desktop type application, not a web page. You can use SquirrelMail or Horde if you want. That doesn't use JS. But why should a project that on the front page has a requirement as JS degrade for those that can't handle it. Not only would it delay the final release date, but essentially you'd have another webmail client, and not the one that I think Thomas has envisioned.
Degrading the JS is a good idea for regular web pages, but this is a web application, not a page. This is meant to be interactive and use Web 2.0 technology. I'm sorry, but I don't see why we should stop down to lesser tech just to satisfy a phone.... most of our users will be using computers and can enable JS for their email site.
Ok, I have read all your arguments about the Web 2.0 application style thing and I agree with them.
However, I think it is very short-sighted not to consider accessibility issues. If I were unable to use a mouse due to handicap, I would not be able to use this application fully. All other windows applications are perfectly navigable using only a keyboard.
I know a lot of work has been done around keyboard navigation and I praise that; I am simply pointing out that one obvious omission is that you can't open an e-mail without double clicking on it. (At least I couldn't work out how to do it, but experience tells me that it should be possible using just tabs, arrow keys and pressing enter on the desired message; every other application works that way and so should this)
My suggestion of using hyperlinks seems the most obvious and easiest way of rectifying this problem and, as a plus, I would be able to check my e-mail on my mobile phone.
I wonder if RC will work on your mobile device when there are hyperlinks available since most links are processed by JavaScript and not directly submitted to the server. However, mobile devices have different requirements to the user interface than a "desktop application".
RoundCube is just an IMAP client which accesses an existing account. I'm not sure if RoundCube has to provide a mutable interface that fits every possible browser or HTML interpreter. I would rather suggest to install an alternative webmail service which is specialized to mobile access. If you can't find one, this would probably be a good reason to start an OOS project...
Both services can access the same mailbox and when you're at home you can read your mails using a regular mail client. I guess this is the idea of IMAP and with such a configuration you always have the best possible client to access your mails depending on your currently used device.
I don't want to bend the interface of RoundCube to be compatible with Lynx because this could possibly get into conflict with the high-end-JavaScript-DOM-AJAX-Drag-n-Drop interface for the regular browser.
Regards, Thomas