next »

[davewa]

Coming soon: version Beta-3.  So far, this version has fixed one bug: the periodic redrawing of a chart which is not showing the sky for "now".  I'm hoping to add code to label the cardinal directions around the chart, too.

[davewa]

Got the labels bashed into Beta 3 today.  I want to rub on the code some more over the weekend and see what other bugs I can fix, but expect Beta 3 to be released early next week.

[davewa]

Beta 3 is nearly ready.

Fixes:
#3 -- now reports syntax errors in coordinates or time and reverts to previous values.
#10 -- doesn't recompute the chart if it's not necessary to do so.

New feature:
The chart now has labels for North, East, South and West.

Give me another 24 to 48 hours, guys.

[davewa]

Beta 3 is now available.  I've edited the first message of this thread to include a link.

NEW FEATURES ADDED IN VERSION 0.33:

C.  The chart is now labeled with the cardinal directions.

NEW BUGS INTRODUCED IN THIS VERSION:

none known.

KNOWN BUGS FROM PREVIOUS VERSIONS REMAINING IN THIS ONE:

1. The text color on the toolbar buttons and labels is uncontrolled.  (I don't understand PyGtk very well.)

5. There is NO support for internationalization.  All strings are hard-coded and time values are using US_eng separators (though the timestamp is specified yyyy/mm/dd rather than the natural order).

8. Rotating the screen causes the controls in the toolbars to be truncated/overlapped.  (The map displays properly, though.)

9. Due to limitations of the algorithms, the accuracy suffers if you attempt to plot a chart for a timestamp more than +/- 50 years of the year 2000.


BUGS FIXED FROM PREVIOUS VERSION:

3. The program will not complain (but will also mess up) if you make typos in coordinate or time fields.

The program will now display a message in the style of "xxx was not understood" and revert to the previous entry if you make typos in coordinate or time fields.

10. Even though you have selected "specify" for the time (and therefore the plot will be for the specified time), the update timer runs and causes the plot to be recomputed once per minute.  This is a waste of system resources.

Now, when you select "specify time", the chart is not recomputed.

[wolf9545]

I don't know if anyone else has tried this but after you setup the "whole-sky chart" program the way you want it, night vision etlc, just press the alt button plus the enter button.  This will remove the top bar and just give you the night sky.  It does take a second or two before it does that.

[davewa]

Interesting.  I'll have to try that.  What undoes that "full screen mode"?

[RobotPrimate]

I guess alt+Enter will exit from fullscreen mode as well? I think there should be a View toolbar, with a fullscreen button the same as in other activities, so that it isn't a easter egg and more visible feature.

[wolf9545]

Oops, I forgot to say what undoes the fullscreen.  Yes it is alt+Enter again.  I know this also works in the reader activity and browser.  I have not tried it in other activities.  It is nice because this way you can set everything up the way you want it and make it full screen.  I have not tried letting it sit like that for a while to make sure the screen updates but I know when it fullsreen mode in reader you can move the pages.

[davewa]

That full-screen mode works very nicely with my activity.  Thanks for the tip.

[davewa]

Upcoming changes (these changes will make the program version go from "beta 3" to "release 1"):

I'm going to try to get i18n support into the next round of code changes.  Besides control text, messages and the cardinal directions, what do you think needs translation?

I want to split the catalog and location data off into a separate Python file to make maintainence of the algorithms easier.  But I guess I don't grok "import" too well yet.  So this may or may not happen.  In any event, this would not be visible to the user (except that it's harder to find the right lines to edit if you want your version of the program to always start with your latitude and longitude by default).

Any suggestions?

[BeckyJ]

Dave,

Could you state step by step instructions for installing Star Chart for Linux n00bs?

TIA

[davewa]

Okay.  But no Linux is needed -- just an XO with the Browse Activity pre-installed and a working internet connection.

Assuming you didn't already install the StarChart activity on your XO, the easiest way to install it is to use your XO to browse to the OLPC activities page
http://wiki.laptop.org/go/Activity#Activities_for_download
and locate the StarChart activity description (about halfway down the page).  Click on the link labeled "download".

(Or just download from http://www.wa1gsf.net/downloads/XO/StarChart-4.xo)

Once the file has downloaded, go to your Journal, find the entry for the StarChart download and resume it.  That should install the program.  You'll have to scroll right to see the StarChar's icon -- it's near the end of the list.

Operating instructions are at
http://wiki.laptop.org/go/StarChart

[davewa]

Coming Soon (Beta 4):  I've been working on a bunch of changes that are all pretty much "under the hood".  I've held off distributing anything because I thought Update.1 would happen in April and I didn't want to complicate that roll-out by releasing a new version while OLPC was trying to make "content bundles" for Peru and Nepal.

But Update.1 still hasn't happened and I guess I might just as well go ahead with the next beta.  Anyway, here's what's going into it:

1.  The catalog data gets separated from the code.  This has two effects:  first, it makes the code (and catalogs) easier for me to edit because they're now in separate files; second: it allows for the substitution of the constellation figure data for locales where the culture has grouped the stars differently, which is a localization problem Pootle never thought of.

2. Plotting of deep sky objects is now possible.  At the moment, I have code that will read a supplementary catalog (of the Messier objects) and plot them -- a bit larger than actual size so they're easy to spot -- if they're big and bright enough for a sharp-eyed observer under dark skies.  (This means about half of the 110 Messier objects are plotted.)  The catalog is missing position angle data, so the orientation of these objects may not match their actual orientation.  This catalog won't ship with Beta 4 unless I can add the PA data and code that uses it to correct that defect.

3. The third feature is something I put together to formalize an ad-hoc patch that I previously wrote for my upcoming vacation.  You can define a file "travel.py" which contains a data array of three values: longitude, latitude and local-timezone-offset.  This last is the number of minutes difference between your XO's clock and the time at the location you're traveling to.  When you return home, simply delete the "travel.py" file to restore "here and now" to your home location.

(For example: my XO is configured for Massachusetts -- 72W x 42N and EDT time; to get StarChart to produce local charts for Hawaii, my travel.py file needs to contain this line:
data = [156.5, 21, -360.0]
because Hawaii is six hours behind EDT and at 156.5W by 21N.)

Still to work out:
1. I18n support is still missing.
2. Some good way for the default observatory location to be customized.  (Maybe I'll just create another small python file similar to "travel.py", that contains the default longitude and latitude.)

Anyway, there you have it:  expect these changes to appear in the bundle sometime before the first of June.  Then I probably won't issue another update until August.

[davewa]

Call for Suggestions

Now that I almost have a working program, I'm interested in suggestions as to what it will take, both in additional programming and in creating documentation to use it (e.g.: suggested lesson plans), to make this a world-class tool for teaching astronomy.

Help me take this activity to the next level!

Add your comments at
http://wiki.laptop.org/go/Talk:StarChart
or post them here if you prefer.

[LesleyT]

Today I was looking at star chart with a 4th grader and he was really curious about all of it. 

What I would have loved was:
the labels for the stars and constellations because I'm rusty.
maybe a wikipedia link to the info about the major constellations and the myths behind them?  I was trying to explain Cassopia and Ursa major and minor and it had been so long that I'm sure that I wrong on most of it. 
(and because my school router HATES my xo, downloadable info would help)

Still, it was awesome to see him so interested