AbeAnd Logs February

From Open Source Ecology
Jump to: navigation, search

Sat Feb 25, 2017

I've had a few glitches creating a stable Linux VM install. A VM isn't ideal, but I don't have my personal machine or an extra disk to install to, yet, otherwise I personally have no use for windows. Many people may need or prefer to use a VM since it can make it easier to use Linux on any machine, though. And with advances in hardware running a Live OSE Linux USB stick or SSD should be feasible soon. I think I've gotten past the install problems( and just realized what caused some), which were mostly caused by my attempts to install the VM on a big 128GB fast USB3 stick.

I also noticed some ubuntu system apps seemed less functional than in previous versions. I am still having some graphics issues, which I attribute to the VM and maybe my settings. But, I do not recall having such issues before with VM's so I will try some different versions and settings of the software to find stability. I did choose the latest ubuntu-16.10-desktop-amd64.iso.torrent out of curiosity. I think I'll test the LTS next since that is probably easier to support for an official OSE Linux.

A good place to start for standardizing an OSE Linux distro might be package scripts. Synaptic eases the creation of install scripts for use on other machines. So keeping an updated package list and install script for OSE Linux may be useful for Support and creating custom ISO/USB installations.

I found some guides to try following to create custom Linux ISO/USB installs. Talk:OSE Linux

Thurs Feb 23, 2017

Still working on my Linux install. I hit a few snags with current hardware setup. Hopefully, I can make a trip to get my desktop soon, which will solve most of my issues. In the meantime, a Virtual Box install will have to do. I Looked over the OSE roadmap and milestone charts again in more detail to get a better idea of how I might align some of my plans to coincide with projects that overlap with my current skills and local projects I'd like to work on. My two main focuses are the critical FreeCAD app and the box of Arduino and related hardware ( and relevant software development ) I have sitting on my desk, which I have yet to put to really great use.

Wed Feb 22, 2017

Doing a fresh install of Ubuntu and OSE Linux apps into a VM since I don't have my personal desktop. Getting everything configured and tested should make for some interesting experience.

I ran across an old article on powerpoint abuse when there was a big revolt against them years ago. It reminded of some of the wonderful works of modern art in OSE google presentations. http://www.techrepublic.com/article/10-things-you-should-know-about-powerpoint-abuse/ and more http://fortune.com/2012/06/12/powerpoint-abuse-how-to-kick-the-habit/ While very dense and technically concise and likely to more visual learners like myself I am not always sure they convey information in the most efficient manner. I'd like to make more use of their versatility, but I don't want to fall into any traps.

I was working back over some code and ideas for my long term arduino projects and while working on a FRAM storage library I figured out a number of bugs and made good strides in understanding better classes and arduino style libraries. However, I also found other FRAM libs I had not before and some other information about possible hardware issues. While I am interested in the FRAM storage for potential future projects where small fast resilient memory is needed I have given in to the idea of using micro SD card for my larger weather and climate control needs. With respect to classes and Arduino libraries, I think I have a clearer idea of how code should be broken into small simple parts so it can be reused or adapted to many projects. I also see significant benefits to agile scrum for coding practices. I do think balancing the priority of product delivery over documentation is difficult for open source development especially for an organization like OSE trying to on board more new devs and contributors who need transparency to see the what, how, and why of everything in order to learn the collaborative work style. Or even for end users and new contributors trying to get up to speed using old documentation that may be lacking. I am considering the bigger picture of many small easy to use Arduino libs for many projects. Following the ideas of delivering basic functionality first making the most critical libs for immediate projects and the most broadly useable seem like the top two priorities. I have been slow in attaining my custom green house controller because of not breaking it down enough. Finding overlap with existing libraries and projects is critical for not repeating unnecessary work. Completing small chunks of code features one at a time should lead to a complete functional package faster. Focusing on nonessentials and rare extra code possibilities has left the hardware collecting dust on my desk. I have also been distracted by uncertainties in how to package the equipment into a safe enclosure so I need to drop these concerns and not go back to square one, but re-evaluate piece by piece. Learning to apply scrum should lead to better methodical decisions.

Tues Feb 21, 2017

Reviewing Logs, cleanup, re-evaluating. While I have been busier with local farm issues I have been skimming the recent wiki changes usually daily & occasionally watch Marcin's youtube. I did get through FreeCAD basics, but not as advanced & as much experience with it as I wanted this winter. For local projects when help (WWOOFers) it is more productive to concentrate on local projects since otherwise little seems to get done. I have some designs to work on slowly this year (work shop), which I'd like to use FreeCAD or Sweet Home 3D for designing. A decent sized local workshop with office space will help me dedicate more space and time to work on projects.

And of course, I want to contribute to the new CEB Press version and related long term big picture plan for earth construction systems. In working with spreadsheets too do estimates, while I decided the more advanced spreadsheets functions & scripts generally add too much complexity for less experienced users like myself and may get in the way of productivity. I think there is a lot of interesting analysis to do on the cost benefits of earth construction systems. I have even been thinking of comparing stabilized CEB production to local natural CEB's aka rocks. We have a lot of rocks here that could be gathered and glued together with lime/cement. I have some experience with masonry labor and rock work. CEB's are obviously less human time and energy intensive since it is difficult to design machines/robots to gather rocks. However, getting the right mix of clay, sand, silt, & lime seems more machine and energy intensive than I first thought. As an example, the area here has more silt with some sand in most areas and then we have found some spots where we mine clay for making cob and CEB's. It can be difficult to evaluate and estimate quantities ratios of natural earth aggregates and then continuously monitor the mix quality with samples.

I see OSE has a possible brick making facility planned and I am curious how this might develop. Generally, a major advantage of earth building is local materials and minimizing transport of materials. Having all the machines tractor pulverizer, mixer, and press easily mobile and streamlining their use process for any work site seems more in line with long-term goals. I do like the idea of a central spot for brick production since many sites may need to haul in some external aggregate anyway and it could make it easier to use solar PV and electric motors in the future since PV is usually fixed and a PV trailer would not likely be sufficient power to run multiple machines simultaneously.

While I decided making google docs and spreadsheets more complex is not generally worth the cost for benefits I think more discussion and clearer guidelines for standard use practices will benefit collaboration. The many different software tools and web portals used by OSE currently or previously can be confusing and jarring in some cases if integration in the wiki is poor or they are separate websites altogether. I like the looks of the OSE Linux distro. I have installed and looked at most of the apps, but need much more practice with all and Linux in general. While an ISO is a necessary static starting point I think many people will use it in a VM since for many users it is a less compatible addition to other software and platforms they may be currently required to use. I am not an IT or VM expert, but I am aware there may be a number of useful Virtual Machine software platform functions that may be useful for collaborative work environments and this will only expand in the future. Syncing VM's in a distributed P2P cloud fashion could be very useful for sharing OSE working data. The other side could be central servers with users logged into VM's remotely so all work is in the cloud from the start. Or ideally some combo of the two models.

I have also recently been reviewing agile scrum because I still don't think I am familiar enough with the working process. Getting more Human Resources always seems to be the issue for OSE so while having simple processes and protocols may not be possible good thorough training information seems most critical so that interested contributors can gradually move up the chain to teams etc. Also having more contributors than team members seems the most likely case this early in the development of distributed open source hardware as well as the most likely case when it is more popular. Many small contributions by non-experts may do more than teams eventually, but the software and management of that scale is currently daunting. So right now small teams and as many contributors as possible seems the most likely state of operations. The learning curve even on complex software isn't that hard for most people if it is consistently used and training systems gradually move users up to more advanced tiers making them able to contribute more and get positive feedback. Feedback and discussion seem critical I found a google doc the other day that had a list of good ideas about planning and discussion, but it seems to have been moved somewhere unsearchable now probably do to editing. I am starting to think that using the wiki is more critical than real time tools like google hangouts chat/audio/video. I have noticed the wiki discussion pages don't get used. This is unfortunate as they are ideal for documenting ongoing prototype decision making since they could log and version information that may not be ideal to clutter the main pages with. Often decisions and why they are made are critical to open source development because someone else may need to make changes that better suit their particular environment or usage case and may need to understand the decision chain of why something was done and how it may affect other decisions, parts et cetera. While nice planning software with graphical decision trees might be nice just documenting more discussion on the wiki would be a good start.