Re-thinking Math

KDE algebraic calculator with visualization in 2 and 3D.
KDE algebraic calculator with visualization in 2 and 3D.

This is an idea and argument that have been prominent meeting with other geeks and people that have had certain success in science. The teaching and use of math in today’s society is pretty low. People complain that people on the street is in need of a calculator even for simple calculations. Also the lack of a mathematical brain, makes people less able to solve everyday problems.

This come in a time that I am teaching someone close to me, pre-calculus as I get the chance to remember some of the things that at the time, seemed so confusing. The funny part is that even thought I believe they are still somewhat confusing, the way the math is being taught also tells me that many of the issues with math depends on the weak links in the chain of understanding the process.

These holes in the process make people less enthusiastic about math. Math problems usually were mainly about the process of solving issues through several steps of calculation. Miss one of these steps and the result will most certainly be wrong. Of course this is nothing new, we can see all these with simple math examples like making a double digit sum, if the first one is wrong, the result will also be wrong.

My proposal is to re-think the problem and maybe take it from a higher level view. Instead of focusing in the process, really target the result. For that, we should let some of the things that stop us from doing that, example, the process. Learning to get results and positive results might be a better motivator than the constant failure of figuring out through quirked processes. Another major idea we should get over  is rethink the whole technology dependency. In all honestly we are dependent of technology for almost everything, transportation, cooking, shitting, entertaining etc. So I don’t see anything wrong with having to depend on technology to calculate. After all, thats usually what technology really has taken over.

However the answer is not just to use a calculator and be done with it. but actually excel in calculating things. That means not only really learning how to use a calculator but other tools like spreadsheets, and of course advanced mathematical models. This basically offer a whole new set of challenges, because we would focus on conceptualizing the mathematical formulas and types of calculations technology offers us.

As a Linux user is really amazing how much mathematic-centric software is there available for every distro. From 3D graphical calculators like Kalgebra to programing language modules ready to be used, like Python’s Numpy. If you are not a Linux user and all you have is Excel, you can spend a good 2 month studying the functions including in the package, as an OpenOffice developer I can give you a hint to see the ones in OpenOffice Calc here. You can see how many of these you have no idea how to use and probably will say that you have no idea why would you ever need it, and that’s the third part of the problem. Inheriting a mathematical education that function in the terms of concepts, there could be a lot of time devoted to inserting the concepts into everyday life.

Sure, you aren’t and probably won’t be an engineer, but at the same time, there should be more people with an engineer means so that engineers can actually move up the scale to think more advanced areas of the knowledge in math for calculating things. In all, I would prefer lessons like Khan Academy to devote to understanding a calculator, drill on those Calc/Excel functions and also more programmatic lessons whenever on Python or Javascript.

In the end, society itself would be able to be less intimidated by math as well as using more math into their daily life. Being able to use a larger percentage of their calculators will make people perform more complex calculations and eventually demand a more calculative data, like stats, analytics and such.

Installing OpenERP 6.1 and 7.0 on the same Linux Box

openerp

After a few rants and raves thrown at the #openobject channel on IRC about why Ubuntu is not the only Linux in time. I was able to get solved some of the issues involved with installing different OpenERP instances and also remember some of the steps that were required to install a general OpenERP system.

First remember that most of the packages will come from PIP but in case some of the packages got issues is important to rely on your distro packages.

The first pain point could be setting up the postgresql database. I experience some issues trying to run postgresql 9.0 and downgrade to 8.5. So first lesson learned is that both database work fine with OpenERP.

Another point is to watch out for the psycopg2 module, in my environment it was best to get it from my repo, since I use mandriva, I need to do:

urpmi python-psycopg2 python-imaging python-pychart

This allowed me to get the openerp-server server running. Another key thing here is to remember to open different ports, so is important to open like this:

openerp6/openerp-server.py –xmlrpc-port=1800
openerp7/openerp-server.py –xmlrpc-port=8059

This is important to get the correct applications running on the different ports. You will end up with localhost:1800 for 6.1 and localhost:8059 for 7.0.

One important thing regarding postgresql is that I wasn’t able to start the services as I should so instead I have to run the server manually doing the following command:

postgres -D /var/lib/pgsql/data/

Following the tutorial on installing openerp v6.1 on ubuntu, I should be able to get the services started from init.d.

Tales from the Kidnapped

Dream about kidnapped family

Today dream was weird, I got a gang coming in my house with my family and basically kidnap us in our own house. The odd thing about that, is that we weren’t tortured, or threatened or any hardcore stuff. We only couldn’t leave nor make any phone calls this remind me of a movie: “Firewall

However, as opposed to that movie, the kidnappers were complete technophobes and even if the phone was secured. I could easily chat with the police and the swat team about identifying the kidnappers and family members.

My dream unfortunately didn’t get to see the extraction operation, but the internal negotiation was most of the content of it. It was an exciting dream, and a bit bizarre.

Localizing OpenOffice to spanish with Lokalize

I am Not a fan of localization, is repetitive, boring, and sometimes confusing. However is something that needs to be DONE. With OpenOffice, there are hundreds and thousands of words on the interface, we need a healthy team to be able to do the translation and another team for the quality assurance. Unfortunately the community doesn’t have enough helping hands or simply are not focused on do actual contributions. In this case only the people with experience are left to go into it and do it. This means me, myself and I in charge on localizing the whole review.

OpenOffice community uses a website at the moment using Pootle, which is great. However the server is deadly slow and translating or reviewing the moves is very hard to do. So I eventually realize that the advantages of instant commit and cloud base review.

Pootle on Apache OpenOffice
Translate Apache OpenOffice through the web.

So eventually I start trying with different PO files, at the beginning I tried POEdit which is simple but compliant and very popular. However is somehow too limited, and to be honest, I still dont get how the memory language works that well. So I actually went into trying virtaal, but this wasn’t a very good choice in my Linux distro that created a python error which prevent me from running it. So eventually I went to KDE proposal called: Lokalize

The tool is actually quite nice, KDE has usually been condemned for looking like windows, however I never believe that crap, KDE looks like a power user tool. And that is what I actually like, I can configure it as much as I want, and having shortcut for many different behaviors actually was worth it to learn some of the shortcuts. And the result: It was awesome.

Lokalize PO and XLIF editor
Lokalize screenshot

I was flying through 100 string modules and was able to commit the translated PO really fast, and even the more scary modules where somewhat long to resolved but in the end took me around 10 or 5% of what it would have normally take on the web.

Key shortcuts like move between strings with Pg Up and Pg Down or jump to the next untranslated or fuzzy string with Ctrl+Pg Up/Down.

To copy the content from the original string to the Spanish one using Ctrl+Space and maybe the best thing to do on localize is the capability to do Search / Replace for the terminology.

Lokalize also offers other things that I need to explore like Autocomplete, Glossary, and correctly use the Translation Memory. I should read the documentation (here) and learn about projects and such.

Translation of OpenOffice 4.0 might come with a better quality however, there are still a lot of layers to be able to understand the way we are putting the accelerators (the key selection letters in a word like File), terminology of tools and even evaluating some of the current terms and it’s extension. One that I keep debating is the use of shortcuts for words like Auto<word> vs <word> automaticamente or Re<word>, lke Rename, Rephrase, Refresh etc. Phrasing is so important that maybe the correct Spanish shouldn’t be the priority.

But I think this will be my tool from now on to be able to work with to translate, and maybe doing something else like the wiki, to be able to break down the jobs on each project. We’ll see how it can go from now on.

Workshop on Enterprise solutions at local University

Today I finished a workshop about Enterprise middleware including a whole letter soup like ERP, CRM, BPM, and MRP. The event was lenghty, it took place in the middle of a series of conferences and workshops including others like videogame development etc.

The Enterpise solution workshop was part of the full course that I have built for private offerings. The course take into account the implementation and overview of an array of enterprise solutions that are under an open source license. The solutions include software like OpenERP, OpenBravo, WebERP, and other solutions like SugarCRM and ProcessMaker.

The workshop was originally planned to have the simulation of a full company interacting between departments where the participant will be part of a department of this virtual company and will execute process such as Sales Orders, Manufacturing orders and finally supply chain.

Process_flow_in_openerp

 Unfortunately several issues like lack of prepareness due to an last minute turnout made me believe just canceling the Workshop. This made me not do the homework needed for the excercises. Then a high level of tiny tasks acumulated to be a bit pressured, setting the virtual machine to make it able for the local PCs to access the data through the network. There also needed to improvise while some of the tasks like adding the users to the system needed to take place.

Dscf0090

Eventually we decided to slowly give some tasks while the workshop really became a conference. As I went on explaining what is the concept of ERP, how to look at it and compare it with many of the real examples. The sad story of internal developed processes where programmers virtually developed a monster of an application that is both new and obsolete and is not well documented, and not compatible with what the next developer might think of it. Eventually becoming a frankenstein applications or on more coloquial terms, a kludge.

Fortunately I was able to put enough theory that even if the assistants didn’t get to play with the application as originally intended. They manage to do the following tasks:

  1. Login the system
  2. Access all the ERP modules
  3. Raise a purchase order
  4. Check the visualization graph
  5. Identify anomalities in the orders

 One of the common hickups some users experience was they werent able to finalize the purchase. This because they needed aditional requisites to go to the final stage of the sales flow. Another hickup was correctly selecting the costumer with the category of the request. This mean that as they fill out the sales order, the product belonged to the wrong category. 

Purchase_form_tab2

However they were able to print the order into a PDF making it good enough for the nature of the excercise. One second hickup wasn’t able to find how to mark the order as Recieved. Hopefully I will get more assistance as I investigate the nature of the process.

After the excercise was done we saw the graph change dramatically not only in the ammount of orders but also on the graph and the impact on the MRP (Manufacturing Resource Planning), obviously demanding more supplies to procurement which needed to approved.  The sudden increase lead to display an impressive graph based on the different category of products making it interesting to find why other products was so high than the rest.

Here it led us to view a ridiculous number that has skyrocket the ammount of money needed for such purchase. Eventually we found an order for many thousands of Ice cream which was funny and insightful since it allow us to look for the responsible of such purchase and get to recognized the power of having processes that can be easily audited.

The next day I was able to move on from OpenERP to other ERP systems like WebERP which was a much simpler ERP to understand as well as OpenBravo which was on the opposite side of the spectrum a way more complete software.

During the session we got to spend time between the diverse applications and be able to compare them. Unfortunately for us WebERP wasn’t as easy to view thanks for the college proxy or the demo was broken at the moment.

I also showed a youtube video about the story of OpenBravo as a case study to see how companies get builted and how it can be done in open source.

In the end the assistants of the conference got what they came for and more. I most say is not bad for a first version of the project and also give me a better view how to conduct these activities. I also got to recieve my activity and people seemed to have enjoyed the conference.

Dscf0074

More pictures from the workshop and the event in general include the following:

Ucaribe001Ucaribe002Ucaribe003Ucaribe004Ucaribe005Ucaribe006Ucaribe008Ucaribe009Ucaribe010Ucaribe007Ucaribe011Ucaribe012Ucaribe013Ucaribe014

 

This vim guy made me feel like a rook

This week I was able to watch a great talk by a VIM guy who went to town on how to use VIM. I consider myself a GNU/Linux power user. Experienced enough to say GNU before Linux and know enough to watch my daily Linux news, and weekly FLOSS podcasts.

However vim is one of those tools that I use everyday but I always feel guilty how lazy I am to not try some of the zillion of keystrokes like many of the ones showed on this tutorial. I found tons of keystrokes that I could very well use on my file editor.

So here is a list that clearly got my attention:

  • w for word advancement
  • a to add a new character after the cursos instead o at the current place like i
  • e: to eliminate as opposed to x
  • f: to find something within the line, this is awesome
  • .: to repeat the process recorded
  • c: for copy
  • and so much more

I end up wanting a part two fof this tutorial and download it for later reference.

Starting early, FLISOL 2012

Flisol2012

So this is a new year and I hope that we can get the wheels rolling about FLISOL. This year FLISOL will be happening on the 28 of April which give us room to perform. I hope that FLISOL this year can help me get even more exposure and make it able for everyone to participate.

The event consist on having many more people involved and have a good place to share the knowledge on the topics. But also to have a more long term vision about the project.

Looking forward to 2013 from now is something that I would love to do, and being able to perform a national event with stronger presence will make it much better not just for the event but for open source in the country as a whole.

I think that the communication channels need to get off the internet a bit and more into the mainstream so that more people are able to contribute for real.

This means that more talks about how to contribute would be a good start to get things on a higher level of deployment and awareness.

Hope to have a great celebration this year.