do you need write access?

Ben Cooksley sourtooth at
Thu Mar 8 18:53:05 UTC 2018

On 8/03/2018 10:47 PM, "Jaroslaw Staniek" <staniek at> wrote:

On 8 March 2018 at 09:51, Clemens Toennies <starbuck at>

> On 08.03.2018 08:44, Boudewijn Rempt wrote:
>> On Thursday, 8 March 2018 07:57:27 CET Ben Cooksley wrote:
>>> On Thu, Mar 8, 2018 at 12:42 AM, Boudewijn Rempt <boud at>
>>> wrote:
>>>> On Wednesday, 7 March 2018 12:31:21 CET Sebastian Kügler wrote:
>>>>> Hi all,
>>>>> We have been working on a modernized website and backend for
>>>>> The new site will do away with the old PHP custom CMS and will run
>>>>> wordpress instead.
>>>> Does that mean we'll lose our history, just like history
>>>> from
>>>> the php times is only in subversion anymore and the wordpress (or
>>>> whatever it was) content is completely gone?
>>> The current content of will still remain in Subversion at
>>> the very least.
>>>  From my understanding the plan Sebas has mentioned envisages a
>>> temporary hostname being kept around until a full
>>> migration is completed.
>> So we _will_ lose proper access to KDE's release history. That's not good.
>> Pages like and everything it links to
>> should be kept up in eternity.
> As far as i can see they will be all available, as they will be imported:

That's very good the content is there. ​Question I would have to the www
community, are we going to keep exact the same URLs?​
This it kind of important key since we link heavily, typically between blog
entry and announcement for example.
I've found our blog content, including or forums having dead
links to

Technically, maybe URL rewriting would help for URLs that were in
standardized form.

Example of as topic harder to handle but possible: currently it
100% redirects to but it could redirect from to maybe or Or rewrite the URL so would not redirect at all (it depends how we want to have it

Example link: to from the
Missing. So relevant announcements for the history of computing, where we
have a place: software for Nokia n900... it's missing so also Wikipedia
references would be dead. The Wiki must to dig in now (see "KOffice
1.6 Released" at

Sibling topic is: dead URLs to relatively new software such as (link from If we have no
resources to share the old files (and mirrors refuse that), maybe we could
at least have a landing page that lists possible options how to find the
archived tarballs?

In relation to tarballs, you will find most of these in the Attic/
directory of

After a while we move things there as we have an agreement with the mirrors
that only carry stable that it will be kept to a certain maximum size.

Even more trouble is missing old image resources but this is lesson to
learn, we still have no ultimate solution IMHO, hash filesbased URLs get
created in and this seems to be fragile approach. I'd think
about at the topic as a matter of Freedom in content area. If we're not
solid here, what can we tell to next generation of contributors that say
"Facebook and YouTube just work".

Then we lost image content on blogs, even image tags seem to be somethow
lost: example, archived
No images, no context... :(

Older posts like this can unfortunately be broken with platform migrations
I'm afraid. That particular post has been through two major Drupal upgrades
at least, along with a change in the domain.

While we do our best it is unavoidable that some loss will occur with any

​I see the task is not light but preserving KDE history is worth investment
maybe even financial. Blogs are to me part of KDE's identity.
Good part of it is done nicely in the form of preserving the KDE 1.0 / 2.0

regards, Jaroslaw Staniek


: A world-wide network of software engineers, artists, writers, translators
: and facilitators committed to Free Software development -
: A visual database apps builder -
Qt Certified Specialist:
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <>

More information about the kde-community mailing list