Enhancements:
– New update notification detection

Bugfixes:
– Fixed memory leaks
– Fixed possible crash on startup
– Jabber: Fixed SASL authentication
– MSN: Fixed MSN contact search
– MSN: Disabled server list options when MSN not logged-in
– MSN: Added removal of new style for MSN color codes in nicks and status messages
– MSN: Fixed infinite loop on authorization failure

Download

• Installer: Unicode | ANSI
• Zip: Unicode | ANSI
• Contrib: Unicode | ANSI
• Debug Symbols: Unicode | ANSI
• Discussion

Enhancements:
– Performance improvements with network logging disabled
– AIM: Added send directory ability
– MSN: Added ability to transfer all files in a directory
– MSN: Use protocol username for chat names in contact list

Bugfixes:
– Fixed possible crash in the file transfer window while scrolling
– Fixed file transfer resume
– AIM: AIM 7 is now detected
– AIM: Improved file transfer support
– Jabber: Ignored the “Manual host” setting during the account registration
– MSN: Fixed chats with IM Control users
– MSN: Auth request reason is now displayed
– MSN: Fixed deletion of contact from Hotmail address book
– MSN: Fixed avatar sending on chat session

Download

• Installer: Unicode | ANSI
• Zip: Unicode | ANSI
• Contrib: Unicode | ANSI
• Debug Symbols: Unicode | ANSI
• Discussion

- Auto away did not respect locked protocols
- Fixed possible file transfer crashes with Windows 9x
- Fixed deleting accounts for not loaded protocols
- Fixed minor selection issues with the profile manager
- ICQ: Incorrect behavior of error messages when popups were unloaded
- ICQ: Improved file transfer resume
- ICQ: Improved status message handling
- ICQ: Improved changeinfo module accessibility
- MSN: Fixed address book migration from 0.7
- MSN: Fixed phones in the contact list
- MSN: Fixed removing IM contacts
- MSN: Fixed blocking contacts not on contact list

Download

• Installer: Unicode | ANSI
• Zip: Unicode | ANSI
• Contrib: Unicode | ANSI
• Debug Symbols: Unicode | ANSI
• Discussion

- Gadu-Gadu: Fix possible hang during login
- Jabber: Handle realms correctly in Digest-MD5 SASL

Download

• Installer: Unicode | ANSI
• Zip: Unicode | ANSI
• Contrib: Unicode | ANSI
• Debug Symbols: Unicode | ANSI
• Discussion

Enhancements:
- Gadu-Gadu: Added option: ‘Show links from unknown contacts’
- Yahoo: Improved login error detection

Bugfixes:
- Installer uses safer method of previous installation choices detection
- AIM: Fixed mail notifications
- AIM: Filenames were not correct on file receive
- AIM: Fixed setting status message to not supported status
- Gadu-Gadu: Icons were not installed correctly
- Gadu-Gadu: Fixed crash opening new conference
- Gadu-Gadu: Removed ‘Show contacts not on my list having me on their list’ option due to the server changes
- ICQ: Advanced search was not working in some cases
- Jabber: Ignored Google Talk custom domains in account manager
- Jabber: Receiving messages stops if tag in response is incomplete
- Jabber: Group chat invites didn’t process non-legacy invites
- Jabber: Fix for rare crash
- MSN: Fixed setting status message to not supported status

Download

• Installer: Unicode | ANSI
• Zip: Unicode | ANSI
• Contrib: Unicode | ANSI
• Debug Symbols: Unicode | ANSI
• Discussion

While it only offers a small sample size on 2 questions, I believe the underlying desire to seek out and understand information about individuals and groups is something we can all relate to.

Fast-forward to the software development context, and the road gets very muddy indeed. There is and always has been a fragile but symbiotic relationship between developers and users. Developers want to exercise their creative and technical abilities to produce something they can be proud of, and users want to be empowered with tools that work with their goals in mind and take into account their individuality.

Why am I babbling on about intangible and nebulous words like empowered, individual and creative? You can’t do these things in any wholesome capacity without KNOWING the individual.

What has all of this got to do with Miranda? Here’s the punchline. We don’t know our users. We think we know our users.

Here is where is it gets hard. How do we move forward? Traditionally and still today there exists 3 main methods; Guesswork, Surveys and Software.

Guesswork – This is the defacto standard. It consists of anecdotal, intuitive feelings about the representative needs of the users. It has brought us a long way, but it suffers from some big problems. It is haphazard, fragmented and not quantitative.

Surveys – This is a method used to add weight and credibility to an argument. They can be very useful if structured correctly, interpreted within context and from an unbiased viewpoint. It suffers some problems too, including small sample sizes, misinterpretation and false inferences.

Software – This is where the software gathers key peices of information from a user and is put into a database where it can be sorted, grouped and aggregated. It suffers from some major roadblocks. Scope, Privacy and Trust.

The following issues come up when looking at the software approach to knowing your users:

• Purpose – What is all of this for?
• Transparency – How and why is it being collected?
• Incentive – What’s in it for me?
• Permission – Should I be asked to participate?
• Notification – Should I be told when I’m participating?
• Opt-in/out – Is it on or off by default?
• Scope – What information is being collected?
• Relevance – Is the information related to the purpose?
• Interpretation – How and what will it be used for?
• Feedback – Will you keep me updated?

I had a wonderfully charged discussion with a project team member I have a lot of respect for, and it became quickly and abundantly clear that this issue has some very deep and important roots that need to be understood and addressed first before any ambitious knowledge project is even tabled.

My motivation is to see Miranda the project and the software grow and evolve positively. I believe there is immense value in our users and contributors who would like to see the same. I believe that equipping ourselves with knowledge is a key component of that growth.

I welcome constructive feedback and critical thinking from the community about why things are important to them, and what we can do to make Miranda as a whole a better experience.

Edit: A couple of people have mentioned that the post left them baffled or a little unclear in it’s purpose. This was half intentional, and half indicative of how raw the thought process was when posting.

My thesis is that knowledge is critical, that we don’t have the amount or quality necessary to base informed decisions on, and that I’d (we’d) like to do something about it. Sharing thoughts and promoting discussion was a first step.

1) The goal of ultimate flexibility / customisation

One of the fundamental design goals of the Miranda project is to empower our users with a piece of software that can be modified to be as unique as the person using it, and it is a known fact that a large part of our userbase are what one can only describe as very technically savvy power users. This goal has meant the evolution of an addon ecosystem and community that is rightly the envy of many other projects.

2) Documentation & Guidelines

These two things are critical in the maintenance and growth of software projects in general, and yet they are often left behind as second-class citizens. Without documented guidelines and resources for our developers, we are destined to make the same mistakes, continually chase our tails and fumble in the dark.

On a positive note, two members of our community have been working for some time on potential solutions, and after tweeting about it yesterday (http://twitter.com/miranda_im) I’ve received a couple of requests from users to upload some screenshots.

I wanted to add some context and background to the problem before I did that, and hopefully generate some excitement around what some of our plans are in the near future. On that note, here are two screenshots that are teasers to ideas that FYR and Nullbie have been playing with. Great work guys :]

FYR’s concept is a modernisation on the existing form, moves the plugin filter control, and adds a option searching function.

Options Concept #1 (FYR)

Nullbie’s concept is quite a change in strategy, introducing ideas like contextual grouping and a simplification of categories and minimising overly complex hierarchies and nested groups/trees

Options Concept #2 (Nullbie)

I find both very interesting and full of potential and would love to hear from anyone who’s interested in providing feedback about whats important to them and why, or anyone who’d like to help with coming up with guidelines, conventions or design patterns for elegent and usable user interfaces.

For more information on how you can contribute to the Miranda project, see our How to help wiki article, or come and have a chat with us on IRC