[Kst] Re: PR proposal for 2.0.3

Barth Netterfield netterfield at astro.utoronto.ca
Sat Mar 26 23:15:36 CET 2011

This is great Nicolas!

Comments embedded below:

On Sat, Mar 26, 2011 at 3:01 PM, Nicolas Brisset
<nicolas.brisset at free.fr> wrote:
> My plan for this communication effort would be:
> - contact Stuart Jarvis to do an interview and publish something on KDE Dot
> News (he already offered to do that a little while ago, and Barth could
> certainly tell a couple of nice stories like committing from the South
> Pole!)

I'll try to bounce back words to him as quickly as possible.

> [...]
> - publish the update on kde-apps.org (right now the latest version there is
> 2.0.0 beta1!!!)

I got the account switched over to me, so I will upload once we release.

> - I have started contacted scidavis and QtGrace developers to see if there
> is interest for some cooperation. I already got some feedback, but as hard
> to believe as it may sound (for people using or working on open-source,
> Qt-based plotting tools) they don't know kst at all, so I guess it's going
> to take some time to identify what cooperation could mean, as they will need
> to try kst - just as we may need to test their tools. I'd like to emphasize
> that cooperation does not necessarily mean dumping some of the tools (even
> though it's always a possibility), but maybe some shared libs or plugins
> would already be interesting. In any case it could be interesting to get the
> information of the new release posted there too, provided that it is not
> seen as an offense by the respective devs

I think there may be capability overlap between the various programs,
but kst has as
its first priorities:
  -fast exploration of real time data
  -fast exploration of massive data sets
Other capabilities, like data analysis and publication quality plots
take a secondary
role, where these other project probably excel in them....

> - post the still-to-be-done screencasts on youtube, as even if the quality
> is not what I'd like for the ones which will be linked to the website, the
> audience there is probably greater and it can certainly help the google
> ranking of kst, which is not in the first page right now when you look for
> "free plotting tool", "free plot tool", or "cross-platform open-source plot
> tool"

excellent.  Can you let us see a draft?

> - pimp up the wikipedia page(s) somehow linked with kst
> - any other good idea you may suggest :-)


> ***********************************************
> Kst 2.0.3 released March 29th 2011
> The Kst team is happy to announce the immediate availability of Kst 2.0.3,
> the not yet so well-known but outstanding open-source, cross-platform 2D

delete "the not yet well-known but" and replace with "an"

> plotting tool. Packages for Windows, various Linux distributions and Mac OS
> X are available, as well as the complete source code and cmake-based build
> files.
> A more detailed presentation of Kst can be found on the completely revamped
> web page at http://kst-plot.kde.org, which provides detailed screenshots,
> video tutorials and all the useful download links.
> Kst2 is a major rewrite to Qt4 of the long-existing, Qt3/KDE3-based Kst1
> code base. Kst2 has not yet reached feature-parity with Kst1, but on the
> other hand Kst2 also offers some features not available in Kst1. Both
> versions are considered stable and can be installed in parallel. This Kst2

as far as I can tell, scripting and vector eps plots are the only two missing
features of note.  The vast majority of people don't need scripting
(in fact I have
not yet seen a real-world use-case for scripting, though I know that they must
exist!) or vector eps plots, so I would prefer if we didn't make the
lack of feature
parity 'front and center.'.  so perhaps change "Kst2 has not yet... in
parallel" with

Kst2 offers many new features compared with kst1, including better meta data
support, improved name handling, and shared axis boxes.  For those who need
the ability to read kst1 files, and scripting (not supported in kst2),
kst1 and kst2
are installable in parallel.

> release has benefited from much more testing than previous 2.x releases and
> is now considered usable for real-life projects, even though there are still
> some rough edges, which are being polished release after release. For a more

delete "even though there are still some rough edges".  There are in
kst1 as well.

> detailed discussion on the differences between Kst1 and Kst2, please refer
> to the bottom of this page.
> Note that in the course of the port to Qt4, it was decided to base the core
> of Kst2 on Qt only for ease of distribution, especially to non-Linux
> platforms. Currently, Kst2 has no dependency on KDE, even though optional
> integration of some KDE features is being worked on to benefit those users
> using the KDE platform.
> Highlights of Kst2 compared with similar plotting tools (like
> xmgrace/QtGrace, qtiplot/SciDavis, LabPlot) are:
> - outstanding performance: curves with millions of points are no problem at
> all!
> - plotting of live streams
> - out-of-the box support for a variety of formats (currently ASCII, netCDF,
> dirfile, Qimage-supported types)
> - very user-friendly and efficient, with a fantastic data import wizard and
> capacity to edit multiple objects at once
> - active community, with over 300 commits since the previous release in
> November and an active mailing list
> - easily expandable for new data formats or data analysis algorithms thanks
> to a plugin-based architecture
> - available on Windows, Linux and Mac OSX
> The most notable improvements brought by Kst 2.0.3 over the 2.0.2 version
> released in November 2010 are:
> - cmake-based build system, with Windows, Mac OS X and Linux installers
> - many improvements in the reading of ASCII files: extensive UI overhaul and
> huge speedup
> - support for metadata added for netCDF files (scalar and string attributes)
> and ASCII (units and some data-file info)
> - UI improvements (new icons, consistency with KDE HIG guidelines, some
> dialogs reworked)
> - lots of bugs and crashes fixed in less-often used parts of the code
> We hope you will enjoy this release and help us further test and improve the
> tool. Kst has been developed over many years by professional developers and
> has reached a high level of maturity, with a very nice, future-proof code
> base. It was recently field-proven during the Boomerang mission in

BLASTpol, not Boomerang.

> Antarctica, during which the probably south-most commit ever made landed in

"possibly southern-most"

> KDE's subversion repository!

> We are aware that Kst is not very well known and generally suffers from a
> lack of awareness, and are seeking to address that with an ambitious
> communication plan, now that we have reached a solid, mass-consumption ready
> release.

I think we can delete this sentence "We are aware ... read release".

>  Please take the time to test kst, report the bugs you may find and
> tell your friends if you like it!
> The Kst Team
> **************************************************
> Sorry for the long mail, but I had lots to say. Now I'm hoping for some
> feedback and suggestion improvements!
> I still have a couple of things to improve on the website and the
> screencasts to prepare. Maybe it would also be nice to give Linux packagers
> some time to prepare packages before announcing the release officially.
> I'm pretty excited to see how this release will be received. I think we're
> now ready to move to the next level, let's see what feedback we get!
> Nicolas
> _______________________________________________
> Kst mailing list
> Kst at kde.org
> https://mail.kde.org/mailman/listinfo/kst

Thanks for doing this Nicolas!

C. Barth Netterfield
University of Toronto

More information about the Kst mailing list