dotMobimobiThinkingmobiForgemobiReadyDeviceAtlasgoMobi
background image

Ready.mobi v1.1

ruadhan's picture
Posted by ruadhan - 08 Jul 2007
Twitter share icon Facebook share icon Google Plus share icon

ready v1.1 screenshotWe've been very busy working on this one, and we can finally unveil Ready.mobi v1.1. Our testing tool has had a major upgrade in version 1.1: no more page-by-page testing! You can now test a whole site with a click of a button. Want to get your hands on this? Then read on...


Before we release this version, we are going to offer the service as a beta trial to a limited number of users. To start using the service:

  • login to your mobiForge user account (or create one), and
  • in your account profile check the "Ready.mobi v1.1 beta user" checkbox (its in the 'Your development' section of your account profile)
  • You can now use your mobiForge account to sign in to the ready.mobi v1.1 beta version at http://beta.ready.mobi

 

But be quick - we can only accommodate a limted number of beta users for now.

 


Posted by ruadhan - 08 Jul 2007

ruadhan's picture

Mobile Web Guy

@rodono
+RuadhanODonoghue

Posted by PaschalNee 6 years ago

Looks good guys - definitely much better that the single page checker that was there.

But it wouldn't be a beta without some bugs! First issue to get you started: the crawler seems to try follow click to call links (an ET phone home type instinct no doubt).

Try a page with the following code:

Contact us: <a href="wtai://wp/mc;0035314806232">0035314806232</a>

and the crawler will try access http://myhostname.mobi/wtai://wp/mc;0035314806232

I'll have a more in depth look when I get the chance.

--
http://mttnow.com/

-- http://mttnow.com/

Posted by ruadhan 6 years ago

Hi Paschal,

Glad you like it so far - looking forward to your feedback!

Ruadhan

Ruadhan O'Donoghue
Editor, mobiForge

Posted by blacknight 6 years ago

Very nice implementation, though it doesn't seem to be crawling any of the internal links on my site :(

Posted by ruadhan 6 years ago

I'll check it out. Note though that the crawl algorithm will only follow links on the same domain.

Ruadhan

Ruadhan O'Donoghue
Editor, mobiForge

Posted by passani 6 years ago

- the old checker would tell me exactly why my code was non-valid. The new one doesn't (either that or I am missing something). That's a pity, because it was a useful feature.

- A new error is generated when there's no .mobi domain. This may be justified politically, but it has not much to do with the correctness of the site being tested.

- Always about markup validity, there are cases where one cannot do without cellpadding and cellspacing in tables, since their CSS counterparts will simply be ignored. The only alternative is to use NON-valid XHTML, no matter if W3C is sad about this or not. A better checker should figure out these cases.

- Google sitemap? what does this have to do with a page validity? why not Yahoo sitemap, then?

Luca Passani

Posted by ruadhan 6 years ago

Thanks for your comments Luca,

- the old checker would tell me exactly why my code was non-valid. The new one doesn't (either that or I am missing something). That's a pity, because it was a useful feature.

You are right, this is missing - we should include it again

- A new error is generated when there's no .mobi domain. This may be justified politically, but it has not much to do with the correctness of the site being tested

This was always reported in the previous version, but it did not contribute to the score, and it still does not in this version. Having a .mobi domain does make your site readily identifiable as being mobile friendly, so there is justification for including it as a (non score-contributing) comment. We didn;t make any conscious decision to change the way this was reported - I'll review it.

- Always about markup validity, there are cases where one cannot do without cellpadding and cellspacing in tables, since their CSS counterparts will simply be ignored. The only alternative is to use NON-valid XHTML, no matter if W3C is sad about this or not. A better checker should figure out these cases.

Agree - there's always room for improvement!

- Google sitemap? what does this have to do with a page validity? why not Yahoo sitemap, then?

Good idea. Bear in mind that anything test included that is not dotmobi compliance- or mobileOk-related is non-scoring. The idea was to check for further "good practices".

Ruadhan

Ruadhan O'Donoghue
Editor, mobiForge

Posted by Vizisoft 6 years ago

Site validates as valid markup on original 1.0 checker, but 1.1 fails due to non-validating XHTML-MP, but no indication as to why ? Problem with the site or the checker - have to assume the checker - how can it be valid / Not valid ?

Posted by ruadhan 6 years ago

This problem was brought to my attention this morning, and has now been resolved. There was an issue accessing the local XHTML MP DTDs, so the validation was bombing out right at the start. Validation should now report same results again for v1.0 and 1.1

Ruadhan

Ruadhan O'Donoghue
Editor, mobiForge

Posted by Vizisoft 6 years ago

Markup now validates, but one other difference is the treatment of images. We use the same image as a small 'icon' on the page multiple times, and the emulators and real phones correctly(?) only request this once per page. However the checker requests this image multiple times, which throws up a large page size error. Again this is different to the original 1.0 checker, and also what we see in real life ?

Posted by ruadhan 6 years ago

Are you sure about this? This is certainly undesirable behaviour, but I was pretty sure that we are only downloading resources once, as browsers do. Can you post the URL you are checking so I can confirm?

Ruadhan O'Donoghue
Editor, mobiForge

Posted by Vizisoft 6 years ago

URL is http://motorhomes.mobi Log extract for the checker and then the emulator is: GET /Default.aspx - 80 - 84.51.242.28 HTTP/1.1 Nokia6680/1.0+(4.04.07)+SymbianOS/8.0+Series60/2.6+Profile/MIDP-2.0+Configuration/CLDC-1.1 - - motorhomes.mobi 200 0 0 4069 330 31 GET /Mobile.css - 80 - 84.51.242.28 HTTP/1.1 Nokia6680/1.0+(4.04.07)+SymbianOS/8.0+Series60/2.6+Profile/MIDP-2.0+Configuration/CLDC-1.1+Java/1.5.0_08 - - motorhomes.mobi 200 0 0 3359 304 31 GET /img/banner.gif - 80 - 84.51.242.28 HTTP/1.1 Nokia6680/1.0+(4.04.07)+SymbianOS/8.0+Series60/2.6+Profile/MIDP-2.0+Configuration/CLDC-1.1+Java/1.5.0_08 - - motorhomes.mobi 200 0 0 2537 308 15 GET /img/mhicon.gif - 80 - 84.51.242.28 HTTP/1.1 Nokia6680/1.0+(4.04.07)+SymbianOS/8.0+Series60/2.6+Profile/MIDP-2.0+Configuration/CLDC-1.1+Java/1.5.0_08 - - motorhomes.mobi 200 0 0 1298 308 15 GET /img/mhicon.gif - 80 - 84.51.242.28 HTTP/1.1 Nokia6680/1.0+(4.04.07)+SymbianOS/8.0+Series60/2.6+Profile/MIDP-2.0+Configuration/CLDC-1.1+Java/1.5.0_08 - - motorhomes.mobi 200 0 0 1298 308 15 GET /img/mhicon.gif - 80 - 84.51.242.28 HTTP/1.1 Nokia6680/1.0+(4.04.07)+SymbianOS/8.0+Series60/2.6+Profile/MIDP-2.0+Configuration/CLDC-1.1+Java/1.5.0_08 - - motorhomes.mobi 200 0 0 1298 308 15

 

[Log file truncated by admin]

Posted by admin 6 years ago

OK, we're caching per page, but not per site. Whats happening here is that each page is being tested independently. So if you use the same image a number of times within a single page, then it will be downloaded only once, but if it is used across different pages, then it is downloaded once per page.

 

On the one hand this gives us worst case scenario: user hits this page directly, without visiting other pages first, so your icon image is downloaded once, and the correct pages size is reported. On the other hand our crawler is not taking into account that you might have already visited another page on the site. The result the crawler is giving is the maximum page size you might receive - during actual browsing, the page-weight might be smaller, but it certainly won't be larger.


Ruadhan

Posted by Vizisoft 6 years ago

Thats not the behaviour we are seeing, even between 1.0 and 1.1.

1.0 reports the page size as "The total download size for this page is 10189 bytes" and 1.1 reports "The total download size for this page is 27988 bytes", quite a difference.

This is the same page, tested as a single page by both checkers. 1.1 is downloading the image file multiple times for a single page, 1.0 does not.
"

Posted by admin 6 years ago

Looks like the resource count is correct but the byte count is wrong - I'll look into it further

 

Ruadhan

Posted by Vizisoft 6 years ago

Looking good - results now compare with the original 1.0 checker. Appreciate the work on the checker, as without it there is no way I could have got ASP.NET to play ball and deliver complaint output. Thanks.

Posted by Sannile 6 years ago

Ronan, hope u are well.

just tried the new version, felt really cool, comparing with the first one i used.

wondering when is the Chinese version come out.haha

Posted by Vizisoft 6 years ago

It looks like the crawler is trying to follow 0121 xxxxxxxx as a page link, converting it into http://yyy.mobi/tel:0121xxxxxxxxx and then obviously getting a page not found error. Is this the expected behaviour of the crawler, or should it not follow tel: links ?

Posted by admin 6 years ago

We have this one on our radar - It should not of course be trying to follow these links.

Posted by Scandiman 6 years ago

Are you still accepting beta testers? I can't find the option in my account profile.

Posted by admin 6 years ago

Its closed for now, but we will be launching very soon