yowko / gource

Automatically exported from code.google.com/p/gource
0 stars 0 forks source link

What is the deeper meaning? #80

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago
Yes, nice program. But what do you use it for? You can look at it till it 
becomes boring, which happens fast. Seems to be inspired by spore.

I am really curious, what is the deeper meaning?

Original issue reported on code.google.com by anton.vo...@googlemail.com on 30 Jul 2010 at 2:20

GoogleCodeExporter commented 9 years ago
[deleted comment]
GoogleCodeExporter commented 9 years ago
I'd *have* to reply to that with an obligatory _yes_.

Focus, anton. Focus.

It's a lovely unique app. Particularly, I like to watch my own history in past 
projects as if I were watching a really, *really* nerd movie.

There is no spoon.

Original comment by leonardo...@gmail.com on 31 Jul 2010 at 4:54

GoogleCodeExporter commented 9 years ago
There is no question which can be answered with yes.

Focus,...leonardo. Focus!

It's a nice unique app. You like to watch your own unimportant history in past 
projects as if you were watching a really, *really* nerd movie.

This is what it is about.

Diagrams are not there as eye candy but as a way to understand data better. I 
don't know how to interpret this movie as no long names are being shown. Some 
points grow and some avatars scurry back and forth.

Eye candy here eye candy there.

What we need is a mixture of usability and eyecandy. Take Sequoia View. This is 
eye candy AND usability.

Original comment by anton.vo...@googlemail.com on 31 Jul 2010 at 9:40

GoogleCodeExporter commented 9 years ago
Gource does show filenames by default, though a lot of users disable this when 
they make videos as it can look cluttered.

Gource is actually an interactive tool for viewing project history. You can 
pause, seek the timeline, and inspect the details of changes. It isn't purely 
for program for producing pretty videos, though that has been a big success!

Were you interested in the history of development of a piece of software, would 
you not find it intuitive to inspect it visually to get a feel for the project: 
The layout. The contributors. Its times of activity and neglect.

This is what I've tried to achieve with Gource.

Cheers

Andrew

Original comment by acaudw...@gmail.com on 31 Jul 2010 at 2:36

GoogleCodeExporter commented 9 years ago
A competent and inspiring answer.

The program is eye-candy at it's best. But I usually loose track. If I put it 
on realtime, there are skippings in the time aswell.

I am using it for web page monitoring with a custom log file.

I created a log-script, which creates lines like that

1280600604|Mozilla-5.0 (compatible, Yahoo! Slurp-3.0, 
http.--help.yahoo.com-help-us-ysearch-slurp)|A|http://xyz.blogspot.com|86bb1d

So there is a user called Mozilla-5.0 (compatible, Yahoo! Slurp-3.0, 
http.--help.yahoo.com-help-us-ysearch-slurp) who scurries back and forth.

How do you interpret that? What do you conclude for your webpage-programming?

All I can conclude is that at times, a lot of users buzz like bees around a 
flower more frequently than at other times(I actually thought about changing 
the user-png to a bee).

Yes it is nice but as time distortion is there and I don't know exactly what 
the bees are flying around(add.: you cannot really track a single bee so it is 
always on your screen) it is just that: nice.

Like watching a bee in the summer surrounding flowers. Like watching a 
relationship of two variables as a stony hill instead of what it is: a 
relationship, which happens to look like a stony hill when being drawn.

This is where strange phrase like "the island of isotops" come from. Of when 
people look at curves of the stock market, attach notes to it and play it like 
a song.

Most people might be satisfied with it: "Wow, so many flying bees" and maybe 
even use it as footage to set to music. It's like visitorville in a sense.

To me games and statistics are different.

I hope you can increase the usability using the lines I wrote.

Original comment by anton.vo...@googlemail.com on 31 Jul 2010 at 7:14

GoogleCodeExporter commented 9 years ago
[deleted comment]
GoogleCodeExporter commented 9 years ago
[deleted comment]
GoogleCodeExporter commented 9 years ago
Your log shows the relationship between the browser User-Agent and particular 
URLs. Since there probably is no relationship, it may not tell you anything. 
Junk in, junk out.

There is a --follow-user option if you want to follow a specific bee.

The time issues you raise are covered by --auto-skip-seconds and 
--max-file-lag. Watching things exactly as they happen (a user adding many 
files all at once) vs one after another in sequence over a short period, is a 
compromise I've made so your brain can process the information. The 
--max-file-lag option let you take it from one extreme to another. This 
definitely one of the less elegant aspects of the program and something I would 
like to improve on.

Cheers

Andrew

Original comment by acaudw...@gmail.com on 1 Aug 2010 at 12:05

GoogleCodeExporter commented 9 years ago
[deleted comment]
GoogleCodeExporter commented 9 years ago
I'd like to understand the visualisation much better. With regard to repo 
history, I'd like to be able to glean some metrics just by watching the video.

I'd love to hear from someone with more experience of using the tool but so far 
I'm kind of thinking that commits where one user is amending many files in 
different branches of the repository suggests such things as highly coupled 
code.

I'd love to be able to put some qualitative comment on the speed at which a 
user zooms around the repo - commits to branches all over the shop - as opposed 
to users concentrated in certain areas of the repo.

I'm going to have a look into the possibility of configuring a different kind 
of display for different kinds of commits (this would obviously be particular 
to what ever conventions were used but we have feature branches, hotfix 
branches etc etc. What would be really cool would be to have the user icon grow 
to something like 5 times its size when a hotfix branch were merged in - being 
able to see how many hotfixes go in would suggest that perhaps more testing 
were required etc etc.

I'd dearly love to hear other peoples ideas on how these can be interpreted - 
to build some kind of profile of characteristics of repositories that others 
could refernce and use to help improve their's. 

Original comment by toonmari...@gmail.com on 14 Mar 2014 at 10:11