[OT]iPhone guidelines
Deco Rior
http://developer.apple.com/iphone/designingcontent.html
The most interesting to me is the mail link.
Deco
--
This list is a free service of LassoSoft: http://www.LassoSoft.com/
Search the list archives: http://www.ListSearch.com/Lasso/Browse/
Manage your subscription: http://www.ListSearch.com/Lasso/
Adam Randall
> Posted today:
>
> http://developer.apple.com/iphone/designingcontent.html
>
> The most interesting to me is the mail link.
>
> Deco
What's interesting about it?
- Mail links open a built-in mail compose sheet.
<a href="mailto:fr...@wwdcdemo.example.com">John Frank</a>
Isn't this what we've been using more than a decade now?
You can also do this to specify the subject, though it's not supported universally:
<a href="mailto:fr...@wwdcdemo.example.com?subject=Foo+Bar">John Frank</a>
Adam.
--
-----------------------------------------------------------------------
Adam Randall http://www.xaren.net
rand...@mcn.org AIM/iChat: blitz574
"Macintosh users are a special case. They care passionately about the
Mac OS and would rewire their own bodies to run on Mac OS X if such a
thing were possible." -- Peter H. Lewis
Eric Browning
:-) Hmm now what to make as a useful app for my new gadget.
--
Eric B
Deco Rior
I've been busy testing AJAX iPhone. ANyone else doing the same?
Deco
Gary Clark
<a href="wtai://wp/mc;0123456789">0123456789</a>
and I thought it was the standard, 'tel' is new to me!
Gary
James Harvard
>A link to a phone number looks like this:
><a href="tel:1-408-555-5555">1-408-555-5555</a>
>
>If you don't wrap phone numbers in a link, Safari automatically converts any number that takes the form of a phone number to a telephone link. If your page contains a series of numbers that could be interpreted as a phone number, but isn't, you need to break up the numbers using span elements, for example.
I think these are poor design choices.
Actually the automatic creation of telephone links is a terrible idea. The idea of iPhone, I thought, was to allow people to browse normal web sites, not just those that have been specifically designed for mobiles. What Apple is saying above is that any web sites that have phone-like numbers have to do extra work to prevent their sites having spurious telephone links introduced. It could be an option that users can trigger to scan a page for numbers, just as on a Nokia mobile you can select the 'use number' option for a text message which scans the message for numbers to call or store.
Conversely sites that do want to accommodate iPhones by having telephone links will then have a whole load of links on it that won't work in other browsers. I can't think of a perfect solution, but I think it would have been better to have used semantic HTML like the hCalendar 'micro-format' http://microformats.org/wiki/hcalendar .
<p>Call me on <span class="telephone">0207 371 2385</span> or on my <abbr class="telephone" title="07973 717 912">mobile</abbr>.</p>
Actually maybe the best thing would be to markup phone numbers according to some standard similar to the above, then have a Javascript that would run on iPhones to turn class="telephone" elements into <a href="tel:"> elements.
James
Jonathan Vanherpe (T & T NV)
>
>
> I think these are poor design choices.
>
> Actually the automatic creation of telephone links is a terrible idea. The idea of iPhone, I thought, was to allow people to browse normal web sites, not just those that have been specifically designed for mobiles. What Apple is saying above is that any web sites that have phone-like numbers have to do extra work to prevent their sites having spurious telephone links introduced. It could be an option that users can trigger to scan a page for numbers, just as on a Nokia mobile you can select the 'use number' option for a text message which scans the message for numbers to call or store.
>
> Conversely sites that do want to accommodate iPhones by having telephone links will then have a whole load of links on it that won't work in other browsers. I can't think of a perfect solution, but I think it would have been better to have used semantic HTML like the hCalendar 'micro-format' http://microformats.org/wiki/hcalendar .
>
> <p>Call me on <span class="telephone">0207 371 2385</span> or on my <abbr class="telephone" title="07973 717 912">mobile</abbr>.</p>
>
> Actually maybe the best thing would be to markup phone numbers according to some standard similar to the above, then have a Javascript that would run on iPhones to turn class="telephone" elements into <a href="tel:"> elements.
>
> James
>
Not to mention that people will try to trick you into clicking (or
however you call the iphone equivalent) a link that goes to a premium
number. I sure hope Apple was smart enough to not have the phone dial
automatically as soon as you click one of those links...
Jonathan
--
Jonathan Vanherpe
Tallieu & Tallieu
jona...@tnt.be
Eric Browning
telephone number. So you get one warning :-)
--
Eric B
--
Stuart Maynard-Keene
- - - - - - - - - - - - - - - -
Kind Regards
Stuart Maynard-Keene
Pageworks
Olivier P. Rouchard
Cup !!!
Kiwies, no hard feelings; you'll have your revanche next September
for the Rugby world cup.
Best regards, Olivier PR
Olivier P. Rouchard
---
o...@copilotpro.com
James Harvard
Does it tell you what number it will be calling with the 'cancel' or 'call' buttons?
<a href="tel:1-900-123-4567">1-800-765-4321</a>
And of course there's JS and HTTP redirects too. Jonathan is right - let's hope Apple gave a lot of thought to security with this!
James
Eric Browning
contained within the href so you can't fudge it by displaying one number
as the link text and a diff number as the number to call.
--
Eric B
Jonathan Vanherpe (T & T NV)
>
> Does it tell you what number it will be calling with the 'cancel' or 'call' buttons?
>
> <a href="tel:1-900-123-4567">1-800-765-4321</a>
>
> And of course there's JS and HTTP redirects too. Jonathan is right - let's hope Apple gave a lot of thought to security with this!
>
> James
>
I'm pretty certain Apple thought hard about it, they're not Microsoft ;)
Actually, the cancel/allow thing kind of reminds me of Vista, I wonder
what they have done to stop you from getting in a loop of cancel/allow
alerts. If you can create a loop of these alerts on a webpage, the
visitor will either have to turn off the phone (I sure hope there's a
button for that on the phone), or just proceed calling the number...
Jonathan
--
Jonathan Vanherpe
Tallieu & Tallieu
jona...@tnt.be
--
Deco Rior
the mobile version.
I am sure Safari could be rigged to prevent this links being active
if this is a problem.
On the other hand this is just a new avenue for malicious activity
which will be exploited to the nth degree. But this is just more
evolution. The most interesting thing so far has been the load on the
AT&T servers.
Although all carriers offer data plans, the overall usage per
customer is low (for many reasons). With the iPhone this should be
reversed with very high usage per customer.
So up to this point has not been worth exploiting mobile users, but
now it is.
However, the benefit for our users will be huge and should outweigh
any negatives.
Deco
Clive Bruton
On 4 Jul 2007, at 15:21, Eric Browning wrote:
> yes it shows the number above the call/cancel buttons. it is the
> number contained within the href so you can't fudge it by
> displaying one number as the link text and a diff number as the
> number to call.
What does it do with: <a href="tel:+442087654321">Call us</a>?
-- Clive
Eric Browning
--
Eric B
Clive Bruton
On 4 Jul 2007, at 16:22, Eric Browning wrote:
> Same thing, just displays the number with the + and cancel/call
> buttons
Sorry, I misread your message earlier:
> yes it shows the number above the call/cancel buttons. it is the
> number contained within the href so you can't fudge it by
> displaying one number as the link text and a diff number as the
> number to call.
If I really got it this time, this is exactly what it does: it
*allows* you to fudge it by displaying one number and dialling
another: ie this:
<a href="tel:+44909654321">+442087654321</a>
Dials +44909654321, rather than the number displayed to the user?
Although, it does ask for confirmation to dial this number.
Adam Randall
> I think he meant the telephone links, that is definitely iPhone
> unique :-) Hmm now what to make as a useful app for my new gadget.
<a href="tel:nnnnnnnnnnnnn">Call me</a>
Has been around since 2000. It's not an Apple creation at all, but Nokia. I found it useful on my Sidekick II when I had that as it allowed me to call things pretty nicely. Apple isn't creating anything here, just using existing technologies.
This is the RFC for tel: based URLs
http://www.faqs.org/rfcs/rfc2806.html
There may be one better than this though floating around.
Adam.
--
-----------------------------------------------------------------------
Adam Randall http://www.xaren.net
rand...@mcn.org AIM/iChat: blitz574
"Macintosh users are a special case. They care passionately about the
Mac OS and would rewire their own bodies to run on Mac OS X if such a
thing were possible." -- Peter H. Lewis
--
Eric Browning
as you are paying attention you will hopefully recognize the pay per
minute numbers for your region. It'd be nice to have the ability in
settings to restrict 900 numbers (US per per minute) numbers as a
parental control through iTunes or via the phone itself with a
password. That would cut down on accidental dialings if you don't watch
what you are calling.
Adam Randall
> On Tue, 03 Jul 2007 21:58:38 -0600, Eric Browning wrote:
>> I think he meant the telephone links, that is definitely iPhone
>> unique :-) Hmm now what to make as a useful app for my new gadget.
>
> <a href="tel:nnnnnnnnnnnnn">Call me</a>
>
> Has been around since 2000. It's not an Apple creation at all, but
> Nokia. I found it useful on my Sidekick II when I had that as it
> allowed me to call things pretty nicely. Apple isn't creating
> anything here, just using existing technologies.
>
> This is the RFC for tel: based URLs
>
> http://www.faqs.org/rfcs/rfc2806.html
>
> There may be one better than this though floating around.
This RFC obsoletes the 2806 one listed above:
http://www.faqs.org/rfcs/rfc3966.html
This is from 2004, and was created by Columbia University
Basically just covers it in more detail from the 2806 one.
If anyone has a mobile phone with a web browser on it, not just an iPhone, you should check if tel: links work as they generally should across the board. I think my T-Mobile MDA (HTC Wizard) does, though I don't have it near me to check ATM.
Deco Rior
about it before, since the browsing experience is so bad!
It seems I was not the only one!
I would really like:
<a href="ical:2006-01-01 06:00:00">6 AM January 1st, 2006</a>
or does this exist already?
Deco
Eric Browning
addition for a quick schedule. Thunderbird however tries to launch iCal
so I guess it is an RFC. Adam will probably dig up an RFC in short order.
>> -----------------------------------------------------------------------
>> Adam Randall http://www.xaren.net
Adam Randall
> alas the iCal link is a no go on the iPhone. That would be a nice
> addition for a quick schedule. Thunderbird however tries to launch
> iCal so I guess it is an RFC. Adam will probably dig up an RFC in
> short order.
>
> Deco Rior wrote:
>> Yes, I realize my Pearl and Treo do that now. I just never thought
>> about it before, since the browsing experience is so bad!
>>
>> It seems I was not the only one!
>>
>> I would really like:
>>
>>
>> <a href="ical:2006-01-01 06:00:00">6 AM January 1st, 2006</a>
>>
>>
>> or does this exist already?
Actually, applications can register a protocol for use on a URL. I don't really understand how it works, but it's per OS. For example:
<a href="aim:blitz574">Contact me through AOL Instant Messenger (iChat)</a>
This will work if you have AIM installed on the machine. I'm not sure if iChat registers it or not, but I wouldn't be surprised. The same goes with the ical link. Technically, the RFC for the iCalender format (yes, there is one, and it was developed by Microsoft and Lotus, and is what iCal uses for it's formatting) doesn't really touch on a URL extension.
Anyway, it may be under a different protocol name too, like cal, or calendar, or may not be registered at all :P
Deco Rior
wow...
So what is the name of the calendar app on the iPhone?
> ----------------------------------------------------------------------
> -
> Adam Randall http://
> www.xaren.net
Mason Miller
Mason
Sent from my iPhone
James Harvard
'tel' links will be seen by all site users, and unless it's an iPhone-only site then most people if they click on these links will find their browser responds with some variation on 'WTF?'.
I suppose that it is useful to have such a link available w
James
At 8:57 am -0700 4/7/07, Adam Randall wrote:
><a href="tel:nnnnnnnnnnnnn">Call me</a>
>
>Has been around since 2000. It's not an Apple creation at all, but Nokia. I found it useful on my Sidekick II when I had that as it allowed me to call things pretty nicely. Apple isn't creating anything here, just using existing technologies.
--
James Harvard
At 7:18 pm +0100 4/7/07, James Harvard wrote:
>OK, but
Wade Maxfield
>that will not work for users accessing the site with something other
>than mobile phone browser.
>
>'tel' links will be seen by all site users, and unless it's an
>iPhone-only site then most people if they click on these links will
>find their browser responds with some variation on 'WTF?'.
>
But this is exactly how all protocols work. Skype, FTP, gopher, mailto, etc.
If I click an ftp:// link and have no ftp program installed (forget
the fact that pretty much every web browser now days can handle ftp)
you get the same effect. It's also how you can add a skype me button
to your contact us web page
<a href="skype://mySkypeNameHere">Talk to me on skype</a>
( http://www.skype.com/share/buttons/advanced.html )
In theory these tel: links could be mapped to your phone dialing app
on your desktop machine (you do have an app that can dial the phone
for you, right?)
- Wade
Clive Bruton
On 5 Jul 2007, at 01:09, Wade Maxfield wrote:
> In theory these tel: links could be mapped to your phone dialing
> app on your desktop machine (you do have an app that can dial the
> phone for you, right?)
I suppose the real interface problem is sometimes not knowing what
particular protocol you are going to be using: just where does that
link lead me: phone/web/ftp/mail...
-- Clive
James Harvard
No.
:-)
Adam Randall
Firefox:
Firefox doesn't know how to open this address, because the protocol (tel) isn't associated with any program.
Safari:
Safari can’t open "tel:9549370049" because Mac OS X doesn’t recognize Internet addresses starting with "tel:".
Adam.
On Wed, 4 Jul 2007 19:18:16 +0100, James Harvard wrote:
> OK, but I still think it's a problem putting links on a web site that
> will not work for users accessing the site with something other than
> mobile phone browser.
>
> 'tel' links will be seen by all site users, and unless it's an
> iPhone-only site then most people if they click on these links will
> find their browser responds with some variation on 'WTF?'.
>
> I suppose that it is useful to have such a link available w
>
> James
>
> At 8:57 am -0700 4/7/07, Adam Randall wrote:
>> <a href="tel:nnnnnnnnnnnnn">Call me</a>
>>
>> Has been around since 2000. It's not an Apple creation at all, but
>> Nokia. I found it useful on my Sidekick II when I had that as it
>> allowed me to call things pretty nicely. Apple isn't creating
>> anything here, just using existing technologies.
--
rand...@mcn.org AIM/iChat: blitz574
"Macintosh users are a special case. They care passionately about the
Mac OS and would rewire their own bodies to run on Mac OS X if such a
thing were possible." -- Peter H. Lewis
Adam Randall
> If you click on a tel: link in a browser that doesn't support it,
> you'll just get an unsupported protocol error:
>
> Firefox:
> Firefox doesn't know how to open this address, because the protocol
> (tel) isn't associated with any program.
>
> Safari:
> Safari can’t open "tel:9549370049" because Mac OS X doesn’t
> recognize Internet addresses starting with "tel:".
>
> Adam.
[
define_tag( 'iPhone_TelLink',
-required = 'value',
-copy,
-optional = 'extra',
-encodenone );
! #value->isa('string') ? #value = string( #value );
return( content_type >> 'iPhone' ?
'<a href="tel:' + #value + '"' + ( local_defined( 'extra' ) ? ' ' + #extra ) + '>' + #value + '</a>' |
#value );
/define_tag;
]
Which you would call like so...
[iPhone_TelLink( '123456789', 'style="color: #f00;"' )]
On an iphone, you'd get an anchor link, with the style setting the color to red, and on a non-iphone you'd get just the number. Of course, this should be expanded so that you can use it with all mobile devices, but this'll solve the problem for the iPhone and everyone else :)
Adam.
Jonathan Vanherpe (T & T NV)
> Right but it will always show the number it is going to call, so as long
> as you are paying attention you will hopefully recognize the pay per
> minute numbers for your region. It'd be nice to have the ability in
> settings to restrict 900 numbers (US per per minute) numbers as a
> parental control through iTunes or via the phone itself with a
> password. That would cut down on accidental dialings if you don't watch
> what you are calling.
>
> Clive Bruton wrote:
>
Is there *any* phone where you can do that? I think the carriers don't
want features like that in phones, since those numbers earn them lots of
money...
--
Jonathan Vanherpe
Tallieu & Tallieu
jona...@tnt.be
--
Jonathan Vanherpe (T & T NV)
>
> On 5 Jul 2007, at 01:09, Wade Maxfield wrote:
>
>
> I suppose the real interface problem is sometimes not knowing what
> particular protocol you are going to be using: just where does that link
> lead me: phone/web/ftp/mail...
>
>
> -- Clive
User agents could show an icon of the app it's going to launch next to
your cursor. There's a Firefox extension that does that (it also warns
you is a link goes to a file instead of a webpage), but somehow a
feature like that doesn't seem to get into the main browser. A good
webmaster should indicate the type of link, though (if it's not a
regular link to a webpage). You can either add an icon manually, or by
analysing each href with javascript after the page was loaded.
Jonathan
--
Jonathan Vanherpe
Tallieu & Tallieu
jona...@tnt.be
--
Clive Bruton
On 5 Jul 2007, at 07:51, Jonathan Vanherpe (T & T NV) wrote:
> Eric Browning wrote:
>> It'd be nice to have the ability in settings to restrict 900
>> numbers (US per per minute) numbers as a parental control through
>> iTunes or via the phone itself with a password. That would cut
>> down on accidental dialings if you don't watch what you are calling.
>
> Is there *any* phone where you can do that? I think the carriers
> don't want features like that in phones, since those numbers earn
> them lots of money...
I've got a Treo, and I'm pretty sure you can put dial restrictions on
it in some form - never really tried it.
-- Clive
Chris Corwin
> Safari on iPhone does not support:
> window.showModalDialog()
> Mouse-over events
> Hover styles
make sure your suckerfish navbars have *link* on the top level.
(i just won a small battle over that, yay!)
- chris
Adam Richardson
On 05/07/2007, at 12:31 AM, Deco Rior wrote:
> Actually, I am not so sure they did! But they do disallow popups
> with the mobile version.
>
> I am sure Safari could be rigged to prevent this links being active
> if this is a problem.
If it's a problem ?
> On the other hand this is just a new avenue for malicious activity
> which will be exploited to the nth degree. But this is just more
> evolution. The most interesting thing so far has been the load on
> the AT&T servers.
>
> Although all carriers offer data plans, the overall usage per
> customer is low (for many reasons). With the iPhone this should be
> reversed with very high usage per customer.
>
> So up to this point has not been worth exploiting mobile users, but
> now it is.
>
> However, the benefit for our users will be huge and should outweigh
> any negatives.
Can you say "slippery slope" ?
http://tinyurl.com/2df43l
reads ...
----
"Baltimore-based Independent Security Evaluators, which tests its
clients' computer security by hacking it, said that three employees
found a way to take control of iPhones through a Wi-Fi link or by
tricking users into going to a web site.
Charles Miller, principal security analyst at the firm, said a
security weakness allowed someone to take control of Apple's Safari
web browser and see other applications on the device at the same
time, which could potentially make users of Macintosh desktop
computers vulnerable to attacks.
"The same problem actually exists on Apple's desktops," Mr Miller
said. But while his firm had identified the risk for both desktops
and phones, it had written only the code necessary to hack into the
iPhone, he said.
The security consultants, who took about a week and a half to work
out the move, said they were able to take control of an iPhone and
make calls or send text messages, as well as access emails,
voicemail, address books and call and web browsing history."
----
If there is one thing that hackers like even more than a vulnerable
browser, it's a vulnerable browser that allows you to take control of
a device which can make paid calls - potentially commercialising a
hack. The bug needs to be fixed very quickly indeed.
Security on internet enabled devices like the iPhone will arguably
become even more important than it is on traditional devices.
- Adam
~~~~
Adam Richardson
Managing Director
Waenick Pty Ltd
Waenick Pty Ltd is a privately owned database development,
data security and online application development consultancy
based in Sydney Australia, with clients in the Australian, US,
and European markets.
We specialise in building database driven business systems for
business websites and internal systems, secure database
engineering, database replication and clustering, high load
webserving strategies, encryption and SSL consulting, and the
development of high security systems handling sensitive data.
We also provide a range of data security services including penetration
testing, application source code audits and network security audits
with full compliance with the remote auditing and testing requirements
of ISO 17799 (BS7799) and ISO 17799-2000 for information security
testing.