[Kolab-devel] The kolab.org website

[Bernhard Reiter]

Hi Paul,

Am Samstag, 13. November 2010 18:39:46 schrieb Paul James Adams:
> Thanks for your input... My thoughts are below.

thanks for your reply, let me briefly respond.

> On Friday, November 12, 2010 12:21:33 pm Bernhard Reiter wrote:
> > I cannot say what "adding website analytics tracking code" would mean.
>
> Piwik works by inserting a small Javascript into your websites main
> template. Each time a page is visited the Javascript is executed sending
> the relevant details of the the website user and what they are up to.

Okay, so it phones home.

> > Personally I do not accept cookies whereever I can and I am getting
> > annoyed by convulted URLs, obstructing the ease of linking to pages.
>
> Piwik neither makes use of cookies (as far as I am aware) or munge URLS. It
> is purely Javascript. It works seemlessly (or not) as long as you have
> Javascript enabled (or not) in your browser.

I also disable Javascript where I can and try to block any phoning home. :)

> > In addition
> > I like personal data privacy which means that I would only track
> > something that is absolutely necessary, which is not much for an
> > information webpage. So I have a tendency against adding "tracking" code
> > in general.
>
> Did you pay that much attention to the Piwik output? There is nothing in
> there which is objectionable. It does not communicate the website user's
> shoe size or sexual preference.

Yes, I've browsed through the output. One point is the felt privacy issues,
I need to trust the javascript that it does not "spy" on me. there have been a 
number of scandals around active contents like javascripts and providers of 
webpages and services actually using the data. I recommend to stay clear of 
this by a wide margin.

> All Piwik does is track where a user comes into the website, where did they
> arrive from, where are they in the world, how long did they hang around
> for, pages they visited etc etc.

That is quite a lot.

> > Looking at the wiki and the webpages, I can see a number of content
> > improvements and I believe that good structure and content is the best
> > somebody can do for their success of a webpage.
>
> I could not agree more. So where do you suggest we start?

I'll make this subject of a separate post if your are interested in it.

> For years now the use of website analytics has been the standard method for
> indicating which parts of a website are not working well. Unless you base
> your restructuring decisions based upon the usage patterns of real users,
> they can only be based upon "intuition" or "pure guess work" as I like to
> call it.

Not using active website analytics does not mean you are "just guessing".
Usually you can make sure to follow general hints, e.g. 
http://www.useit.com/alertbox/9605.html (Top Ten Mistakes in Web Design)
which are based on studies which are more representative than our own 
tracking.

> > Given the limited resources
> > there I would work on the obvious points first and then employ simplified
> > usability tests with people we know
>
> Come on Bernhard, you have got to be kidding me? :-) Do you seriously think
> usability testing is less resource-consuming that adding 10 lines of
> Javascript to the website's main template?

Something that does not need to be done is the most inefficient. (Some claim.)
I believe using web analytics is not necessary and not useful right now, thus 
I believe is highly inefficient. The image risk is also significant as 
explained above.

Beside using the general lists, using a one hour usability test or so can 
already be successful, see 
http://www.useit.com/alertbox/20000319.html (Why You Only Need to Test with 5 
Users)
  As soon as you collect data from a single test user, your insights shoot up 
  and you have already learned almost a third of all there is to know about 
  the usability of the design. The difference between zero and even a little 
  bit of data is astounding.

> We /could/ do usability testing using 10 guinea-pigs and take a few weeks
> over it. Or we could just include the Piwik Javascript and gather the usage
> patterns of 1000s of visitors in a day or so.

We could take the ten often made mistake, tell an intesting story and 
then have one new user of us test it for two hours and work on the list of 
biggest issues we have. Once they are gone, we take another user user.

Best,
Bernhard


-- 
Managing Director - Owner: www.intevation.net       (Free Software Company)
Deputy Coordinator Germany: fsfe.org. Board member: www.kolabsys.com.
Intevation GmbH, Osnabrück, DE; Amtsgericht Osnabrück, HRB 18998
Geschäftsführer Frank Koormann, Bernhard Reiter, Dr. Jan-Oliver Wagner
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: <http://lists.kolab.org/pipermail/devel/attachments/20101126/9edfe9f1/attachment.sig>