Comments in no particular order
More clearly indicate things that things to put in the Eclipse build path
Code was too small on the projector/hard to read
More an explanation of which technology is doing what - still a little abstract
For Beginners
- Pace was a bit too fast
- compounded by network problems
- perhaps a separate track for true beginners
Other venues to use
- Access Grid - video conferencing
- Break the content into smaller segments
- Web conferencing
Comprehensive Reviews
Lisa Wilson
The Simplest Possible Tool - Mike Osterman
Comment - too fast!
I was not able to follow during session as it was going too fast for me, so started reading online info and doing it myself. Got to a point in the instructions where there was code to copy, but no explanation of what I was to do. Got help from Bill Crosbie at that point and was able to finish.
Suggestions: The documentation is great... at first. Please continue updating with step-by-step instructions for us newbies, especially at the later stages as the required steps/tasks are not readily apparent (at least, until they're ingrained.) Include instructions/screen shots for Windows and command line as there are too many assumptions that your audience knows this stuff already.
Putting The Framework to Work - Zack Thomas
Comment - too complicated and too much typing!
The concept of creating files and walking the user through was great, BUT...
The project was still too big if the intent is to type everything - too many classes & components to keep up. If we're going to download this later, why type in front of us? The few hints in troubleshooting were good, but too far in - I had already lost track of what was going on.
Suggestions: There was no conceptual design behind this, i.e., tool design sprung out of your brain wholly intact! Unfortunately, it doesn't work that way for me. What are we doing and why?!
More comment during constructor creation about deciding what functions are needed and maybe what happens when you go back and add things in you forgot in the first iteration.
Finally, there are no step-by-step instructions online for this stage, so I'm lost at this point with no hope of doing this by myself.
Using Java Server Faces - Thomas Amsler
Comment - good job!
This section was better organised and the pace was better too. Speaking voice has improved - not quite so monotone until towards the end. The steps were easier to follow and well documented.
Unfortunately, my build has not recovered from session 2, so I wasn't able to deploy maven successfully. I'm going to have to try this out on my own.
Solid work, good checking with the audience.
Persistence in Sakai - Aaron Zeckoski
Comment - even better!
Clear, concise. Good voice, well-documented online content. Again, since I'm behind, I couldn't try anything during class, but I was able to follow along. Aaron seems comfortable with speaking and his presentation was well organised.
Why are We Here - Bill Crosbie
Good identification of a burnt-out audience. Too bad we didn't get to testing, but I doubt I could have followed anything by this point. Hope to see documentaion on testing though.
OVERALL COMMENTS AND SUGGESTIONS
The biggest stumbling blocks for me were using Subversion and deployment failures in maven. I still don't know how to update files via svn and stopping to troubleshoot means I missed parts of the presentation.
It would be great if the online site were reviewed for consistency. It's tough with various participants to get a common style and some parts are difficult to understand.
Overall - I would give this cafe a thumbs up. As a continuing support process for the community, great strides were made in the right direction. This was a solid attempt. I hope the online material is updated as I plan to run through this again, and I'm sure the community who could not attend will find this resource valuable.
Comments (11)
Jun 02, 2006
Cheryl Johns says:
I was very pleased with the documentation on how to configure your laptop for bo...I was very pleased with the documentation on how to configure your laptop for bootcamp and that was a valuable lesson for me.
The pace, for me as a NEWB, was a little fast because I was trying to make sense of what we were doing as we were doing it and would get lost if I spent a little time trying to absorb each step.
I did things that I really don't know WHY I did them, but know they work. Perhaps a 30,000 foot view of how the applications are created and what pieces do what would help (maybe ya did that and I missed it furiously typing away).
I like the cut/paste idea so we can catch up on getting the code into place, but maybe still watch the instructor type it out during the explanation.
I found it very valuable and hope to be able to review at a later point.
I wasn't really clear on why there were 4 iterations and why you would need them. Is the ultimate goal iteration 4? I wasn't really sure. Were they style preferences? Dunno.
All in all, I think it was a great success and hope to see more of this!
Jun 06, 2006
casey dunn says:
More projectors perhaps on the side? more smoke machines too! I was impressed by...More projectors - perhaps on the side? more smoke machines too!
I was impressed by a couple of little things the folks at JavaOne did in their labs -
Hand outs; I know hardcopy is so passe', but given the limited real estate on ppls laptops...
CDROM/DVD of course materials, staged.
Jun 06, 2006
casey dunn says:
rats, something else; is there a summary of the svn layout outside of the letctu...rats, something else; is there a summary of the svn layout outside of the letcture notes?
some sort of "you are at stage N, use tag Y" map?
I write this as I'm trying to remember what the trunk subdirectories are for and how they may differ from the branches.
Jun 07, 2006
Paul Beckett says:
Firstly I want to say thankyou to everyone involved in organising this event it'...Firstly I want to say thankyou to everyone involved in organising this event - it's great to see such a high commitment to helping other people learn how to develop sakai tools.
As someone who has deployed sakai, but not done any development with it, I found the pace too fast (especially when network problems compounded the time pressures). It seemed to me that some people who had already dabbled with sakai development where comfortable with the pace. Perhaps two 'streams' would be appropriate: one for newbie's (like me) and another for those with a bit of experience? Maybe a 'newbie' session would need 2-3 days to cover all the material at a comfortable pace?
I am looking forward to getting a lot more value from the materials you put together by working through them alone, at my own pace. My only slight concern is whether the documentation without the audio commentary from the session is sufficient?
I noticed, you were recording the sessions with a video camera, is anyone planning to turn this into a series of movies (like Zach's how-to-develop-a-sakai-tool video's). I think that this would be an amazingly useful thing to do. I know it would be a resource I would value.
Shorter Video conferencing / access grid, each covering one iteration might also be very useful. Although with such a large dispersed global community perhaps this is difficult because of time-zones.
Jun 07, 2006
casey dunn says:
Paul, I know these guru's wanted to spread the sessions out over the duration of...Paul, I know these guru's wanted to spread the sessions out over the duration of the conference but it was 'difficult' to schedule.
what do you think that would of been like? would the breaks allow time for you to digest what was going on, or would there of been too much distraction?
Jun 12, 2006
Paul Beckett says:
Casey, I'm not sure if spreading the sessions out would be good or not. I do lik...Casey, I'm not sure if spreading the sessions out would be good or not. I do like the idea of having a big block of time to focus on learning about developing a sakai tool with no other distractions.
On the other hand, gaps between sessions, would create the possibility of people spending time (as you say) to digest what they'd learnt, and also to get up to speed on anything they missed / didn't understand.
If it had been spread out I think I may have had a better chance of catching back up again - having become lost early on: partially because of trying to sort out my network connectivity problems and partly because the pace was too fast for me.
Jun 09, 2006
Lisa Wilson says:
Hi guys, I spent today reviewing content under th Programmer's Cafe and reviewin...Hi guys,
I spent today reviewing content under th Programmer's Cafe and reviewing Zach's first movie. It struck me again how tough this is coming from a Windows rather than a UNIX environment. It's definitely a disadvantage since almost everything is geared for Mac/Linux.
By supplementing with resources like Guide to TortoiseSVN I've been able to download sakai 2.1.2, but I'm not sure what to do because I already have 2.1.1 up.
So, that brings up other questions, like can you do parallel development (e.g., you're working on both your current and future versions) and how do you point Maven, Eclipse, etc. to a new version.
Thanks for all the help you've given so far!
Lisa
Jun 12, 2006
casey dunn says:
Here is what I do but then again you've seen me in action so I'm sure you realiz...Here is what I do - but then again you've seen me in action so I'm sure you realize just how crazy this may be
Using eclipse's workspace capability I keep seperate workspaces for each version. I load those workspaces with (holding breath) svn checkouts from our local repository's branches of offical Sakai release vendor drops.
On Windows I use tortoisesvn to manage directory tree along these lines:
I also use cygwin.
To build I use a set of cygwin shell scripts which drive maven with full command line versions of the build.properties information, tuned for each version (2.1.0, 2.1.2) and each deployment stage (dev, qa, preprod, pilot...)
my small integration team (myself, a QA pern, and our revered sysadmin) uses these same scripts on our varous platforms; windows, mac, solaris, debian... clearly cygwin fills a central role in this scheme.
Jun 12, 2006
Mike Osterman says:
Casey, If you could package these scripts of yours or Confluence (using the curr...Casey,
If you could package these scripts of yours or Confluence (using the current cultural trend of verbifying nouns) these scripts you use, I, and likely others, would be greatly obliged.
I too use the separate workspaces and shell scripts, but it hadn't occured to me to invoke the build.properties values in the shell scripts, despite having seen it before.
While it's not strictly necessary as some have pointed out, I can't emphasize how much Cygwin can make life easier for Windows developers, even outside of Sakai development. It also helps to build some cross-platform shell mojo, which is no bad thing in itself.
Jun 13, 2006
casey dunn says:
I don't know about 'package' but I could provide a snapshot. :) we're Tinkering ...I don't know about 'package' but I could provide a snapshot.
we're Tinkering as we go with these scripts, attempting to be Agile where we can (as opposed to w/in Sakai at our current levels of skill and, well, the rest...)
We've an internal confluence page giving broad strokes. so I can start with that and drill down locally.
Where Confluentially, oh Mike O? Where would Lisa, or any other reasonable person, find them; I defer to your useful momentum.
Jun 13, 2006
Mike Osterman says:
I think the internal confluence page is a good start. We've got to find a home f...I think the internal confluence page is a good start. We've got to find a home for all these useful scripts floating around - perhave a "Tips & Tricks" section of the Development Environment Setup area?
The other thing is that we should probably unlock edits on this space. It made sense as we were trying to refine the pages for the Cafe, but it would sure be more community-minded to open it back up.
On a side note, I'm thinking a "Cygwin from scratch" page would be nice, along with some tips and tricks to show the inherent coolness and value of said environment.