Today I had the pleasure of meeting this man, he is huge in knowledge and human values. He told us his beginnings in business as human needs for getting money to solve humanitarian problems, as well as advice on lessons he learnt over 20 years of business. Apart from metting a very interesting person with great stories about entrepeneurship, I would focus on these points:

Listen to what your users “use”, don’t ask them what they want

This is a nice thinking, specially after metting about A/B testing. I agree with this since generally people will not tell you what they want. Even more, they don’t even know what they want many times. But when they use some service over and over again, then you listened to them, even if they never told you to accomplish this or that.

So feature requests and requirements from partners and users should be put in a can and consider them but no more than your own intuition and the trends in your users: what they use, how they use it, etc…

Skills change over time

He undervalued people resumes, since changes happen fast and what someone did great 2 years ago may not be good to what he needs to do 2 years from now in a new venture or start-up. Also, reward people when they deliver not for having a great resume or cv.

His philosophy of funding businesses and start-up companies can be found on the book he wrote, will have it in my wish list, must be great reading.

Also agree on the bootstrapping as long as you can, and start getting income as soon as you can, forget about investors in eraly stages, stating that we know many sucess stories but not the fail stories.

Ah, by the way, this guy helped to make Xing a great success, investing and delivering a product along with the founder in 90 days. He likes to develop software globally with people in India, Pakistan, Canada, US, and use Ruby On Rails. I think this is ok, but would not focus on language, just development philosophy, does not matter in my opinion which language you develop on, but have a great software development team.

Tweet This Post Plurk This Post Digg This Post Ping This Post

I started to use GoogleApps in the last work project, where everybody is using gmail for corporate mailing as well as google apps, etc… So I dig deeper into google apps to host by domain there and dump the Exchange mail hosting solution I have now. It ended up that instead of paying 10€ a month for exchange+blakberry bed would pay like 4€ a month with GoogleApps. But since this crisis made me very cautious about spending money, I said to myself to search deeper in the net.

Zoho Mail

I reached Zoho, where I could host my own domain mailing at 0€, having IMAP and POP, therefore being able to sync with my iPod, Blackberry, from any browser, etc… Sounded like a great deal. I have 1GB for first 3 users, which is a great proposition for start-ups. Their spam filter is way cool. Believe me I have tons of spam every hour, the zoho spam folder as I am writing has 8,000 messages and that in a few days only. It catches spam pretty well, at least the kind of spam I get. And finally I can send my personal email from Yahoo to the same account with label personal so have all email organized in same place. I get 25GB paying small fee like 3€ when my mail grows which will not happend soon, so I am OK.

Zoho Apps

Now this is the one that rocks. I get Calendar, Tasks, Notes, Forums, Link Management, Contacts, Notebook (MS OnNote style), Wiki, Project Management, Meeting Software, CRM (Like SalesForce), Share App, Invoice App, SpreadSheet (GoogleApps compatible), Planner, Writer, Chat, Marketplace, People (Human Resource Soft), Recruiting Software, Slide Soft, Reporting Software, Remote Desktop, Document Online Disk (1GB) and Creator (Web site creator) totally free. This is really awesome.

Their application are not crappy apps, they are nice applications, usefull for everyday use. And the number increases like monitoring soft, polls, etc… So I say: “This company delivers”. This is the reason why I moved to Zoho and plan to be there. When need more proyect management will pay very low on-demand fees, the same for the other services. So seems to be much better for me that GoogleApps. And as I said, they have compatibility with GoogleApps and their calendar application so no problem with that.

I set up a public calendar that will publish into this blog with the online events I go and public stuff I go, etc…

Also, besides all this I can choose mail account name and not be jalegre8726@gmail.com, which is not very compeling to me.

Tweet This Post Plurk This Post Digg This Post Ping This Post

Content from a search engine is very important to have a quality search service for users. A few months ago, I started to think that unless I had the chance to have funding, I would not be able to compete with the big guys due to lack of infrastructure, etc… The funding failed but seems it is not that critical at all.

Current Data

My startup is a new search engine. I have now like 6 million web pages in spanish. Content is very low with respect to Google and Bing, to name a few. But on testing I did, I have average 50% of the content Google has on first page for popular queries. I assume this is not that bad considering that so far I have no modules to get only authoring or popular content.

Fetching Popular Content Module

I am testing a module that creates a database of all the links outbound from those 6 million web pages (which are reasonable relevant). The db is building now, in a few days will have the results. I assume I would be getting like 10 million links. After that point, B+ will only index content from these web pages, leading to a higher percentage content. Anything above 70% will be good, and figures like 90% would be great. This improve will lead to better search results for users.

Why First-Page Google Is Important?

I am not focusing on giving results on first-page from google, but this factor is important to tell how much relevant content B+ has in spanish. The personalization technology and preferences feeded into B+ will allow nice search quality for communities and groups of people. So first page analysis gives me a pretty good estimate that I have most quality content and lack of funding is not ruining the evolution of this search project.

Soon will publish on the blog results

Tweet This Post Plurk This Post Digg This Post Ping This Post

It was nice to watch this video. Daniel is a great communicator and as allways TED content has great quality. This event was at Oxford, would be great to attend to these events life.

Panda CEO some months ago was talking that stock options and so forth creates merchandisers instead of a good team of talented individuals. I had  my doubts at that time. Watching the video I agree on Dan terms. While working 100% on my start-up the three motivators mentioned in the video got me moving. These three are:

1. Autonomy: Desire to direct our lives.
2. Mastery: Desire to get better and better in something that matters.
3. Purpose: Participate in something larger than ourselves.

I think these three motivators can work for most people. I think that this will depend from person to person the level of motivation for each point. For overall will work for most people. People that value security and belonging would go with #3, active people will love #1, etc…

Working as consultant in software development for clients I found that people I talked about this they were not that much interested in better salary but rather that their opinion counts, they are listened and influence, etc…which would go to #3 becaouse if you are working on a project and have no level of influence then “you can no purpose”, right? becaouse you are not really participating.

Tweet This Post Plurk This Post Digg This Post Ping This Post

I started the open source project simple-saas a few days ago, hosted at Google Code. The idea is to help my start-up processing times and provide a multi-language remote processing system with cloud computing in mind.

• Client/server software with JSON serialization with objects support. We use jsonpickle for this under python. Idea of any client communicate with any server. For example, PHP frontend communicating with a Python or Java server calling a service.
• No need to change anything on existing classes.
• Map&Reduce type of processing with less overhead.
• Objects are mapped in mapping servers running database software OctopusDbm with those mapping being in memory and therefore pretty fast. We mapp against application name, which can be any string you choose, defining domains in string, etc…

Last night I was in process of how to map the services and objects, so came up with the following design to connect to a server:

from simplesaas.client import Service
myService = Service(sHostName, 'module.package.MyClass')
client = myService.getClient(iClientId)
print client.Name

You can add any arguments and keyword arguments to the Service() instantiation and calling of method that the class and method on remote server needs. The module in the server just needs to be in the python path.

and this to connect without having to define a server to connect to:

from simplesaas.cloud import ServiceCloud
myService = ServiceCloud('module.package.MyClass', sApp='myDomain:myApplication')
client = myService.getClient(iClientId)
print client.Name

sApp is a string that can be anything, so we can define just application name or a combination of domain, application, language, etc…

The mapping server will have:

sApp => sHostName

The mapping will be included in mapping servers with a script and calling Octopus data operations.

I thought about allowing possibility of not defining an application and client will connect to same host where client is. But I think is not good design since later on, when that application is taken to a distributed environment then code must be changed. You can define a constant in your environment to define, say, a “main” application with a constant or similar.

There will be a config.py file in the simple-saas package installation where we define:

• Tuple with list of mapping servers
• OctopusDbm user, password, port and environment

You can define just one OctopusDbm server or two which could be replicated.

Let me know what you think about this design, flaws, things should be different, etc…

Tonight will go through the map and reduce functionalities., probably focused on patterns of design. First pattern will be getting a list and each server processing a sublist of that list, being able to make requests at any time and later on getting results.

Tweet This Post Plurk This Post Digg This Post Ping This Post

I used the points accumulated into my Vodafone account and got a new Blackberry with 0€, which is nice. I was using a blackberry 8700 for a long time, so will be nice having camera and wifi on it.

I have an iPod Touch which I use a lot, so guess browsing will still use the iPod, but now I will have camera to upload pics to facebook, etc… as well as video. The device is GPRS/EDGE Quad band which is nice for travelling.

Can sync with my Exchange server and have same things on my iPod and Blackberry, email, contacts, etc… while on Wifi. I do not expect to have a BES as I used to have with my blackberry due to cost and I can get along downloading stuff while on Wifi.

A nice thing is to have the newer blackberry OS and play around with blackberry apps. Besides, having keyboard is nice for writing my long notes as I used to do and sync them.

Why don’t get an iPhone?

Well, first is cost, I would go back to a costly solution like I had with the BES. Second, will not move to Movistar no matter what. And thrird, is nice having two devices, one long battery one and the other fun and short battery. Sometimes happens to me that iPod is out of battery, but still my BB still can use it in Metro for example to have maps, play, etc…

Tweet This Post Plurk This Post Digg This Post Ping This Post

Last weekend I worked on the design of the new modules for OctopusDbm and looks like nice features are coming in soon. The first release was built, 0.6 and soon I will start documenting the current modules and design.

Community

Any open source project needs a community or group of people that like the technology and contribute with ideas, code, testing, analysis, etc… It will be a nice experience for me to learn how to do this. So the following weeks I will start promoting the project and trying to attract lovers for this project. Spread the word if you read this, this tech can be pretty cool.

Dbm

The client/server implementation works in such a way that the code change for applications already running BerkeleyDB software local is minimal. Almost all API methods are implemented in remote server.

Curently you may have:

from bsddb3 import db
Dbm = db.DB()
Dbm.open(sPath, db.DB_BTREE, db.DB_CREATE)
Dbm.put('Name', 'Jorge')
Dbm.close()

And with OctopusDbm:

from octopusdbm import Dbm
dbm = Dbm(hostName, user, password)
dbm.open(sPath, db.DB_BTREE, db.DB_CREATE)
dbm.put('Name', 'Jorge')
dbm.close()

So you basicly can open, put, create cursor, get, get table information, etc… same as you do in local. Keep in mind that for every database methods there is a call to the server. So server connections for open, put and close. If you need better performance we have direct operations where open and close are integrated, like getDirect(), getLIstDirect(), as well as bulk operations like getList() and putList() with tables opened.

No Db Sessions

One of the design points of OctopusDbm is that you don’t need to open connections to database or create sessions before processing operations in database. The idea is that getting and putting data is as simple as getting an HTML page, so simply a request is performed and you get result. This way is easier to connect to 10 or 20 cloud images without pool of connection, etc… Imagine having a pool of 3 connections to 100 cloud images, that would be 300 database connection to manage for every node needing database connectivity to the cloud.

This is of course open to discussion, so place your ideas, etc…

DbmCloud

We introduce this object for access to the cloud. The distribution has been designed so far to e horizontal. That is, table 1 goes to cloud image 1, table 2 to cloud image2, …., table 11 to cloud image 1 when having a cloud with 10 images or virtual machines. This object also manages partitions in horizontal way and while you have an integer primary key or index. So tables being in the cloud must have integers if you want to partition them. If they have string indexes they will not be partitioned so far. I think this is ok for first versions. You can allways pass this DbmCloud, implement your own, participate in the project with your own dessign of cloud distributions and partitions, etc…

It has been decided so far to have an XML file with all schemas, tables, defining number of partitions, number of cloud images organized by domain and application name. For very big systems, they could have different domains and application inside each one with a cloud design. Small apps would have 1 domain and 1 app.

One cool aspect of DbmCloud is that any database object, that is, data table, index file, foreign key will be distributed and queries running on them executed in different cloud images therefore boosting performance, even more when smart partitioning in place.

Models

In order for OctopusDbm be an alternative to SQL databases like MySQL we need a model design that will:

• Allow us to have tabulated data in columns
• Design to make it easy to implement persistent subsystems like Hibernate, JPA, etc…

We thought to do this inside Python similar to Django models, where you define class attributes with classes relative to types and using dictionary class attributes:

oClient = Client(Name='Jorge', Place='Madrid')
oClient.save()
 
oClient = Client.get(Name='Jorge')
oClient.Place = 'Madrid'
oClient.save()

First one creates an object in database system and writes to database in a lazy mode. Second one updates the client object in database. I think the models would support cloud distribution and partitions. You define in the class the attribute types, index names, physical names, etc,….

The primary key in tables can be an id or other value. There will be a type IdGenerator that will generate auto ids for rows, but you could define you own primary key values. When support for cloud and partitions we need an integer primary key for partitions.

First versions will probably lack foreign keys, but later on we can implement foreign keys with model integrity. I thought of Links between tables besides ForeignKey types to execute soft links for example to get literal of an id in a parametric table.

Tweet This Post Plurk This Post Digg This Post Ping This Post

Nice article about taking notes software here:

If You Didn’t Write It Down, It Didn’t Happen — Take Notes

I used OneNote from Microsoft and was nice the visual aspect of the tool, but stopped using it, I guess becaouse I started to take notes on my blackberry, sync with Exchange, etc…

My opinion is that the networking of these tools is very important. I signed up for Evernote but never catched me up enough to feel great about the tool.

I would add Twitter as note taking, many people use Twitter to do that, write down things so he and others can take a look at it later on. And could ot be simpler note taker than Twitter, right?

Tweet This Post Plurk This Post Digg This Post Ping This Post

I tested new Oracle Java edition for Berkeley DB and for my tests the word that best describes the product is FAIL. The conslussion more or less is that Oracle screwed Berkeley software. They have no HASH, only BTREES, many functionality of BerkeleyDB has been removed, etc… They say is BerkeleyDB with Java flavour, they have a weird sense of java humor then.

I was expecting factor of 5X as many blogs suggested that on real applications we have this improvement. And since BeerkeleyDB for me has a lot of memory use and processing as well I was expecting at least 2X factor. I got a 2X factor but on the contrary.

Results

I did test for creating a table of 1 million rows with 128MB memory Cache.

So the refactoring of OctopusDbm, if any, will go through C++ inside Linux. I downloaded the Linux Tools for Eclipse and will start playing with those when I have some time.

It seems that Python with C modules work great.

[Update Feb 22 2010]

The tests were performed 5 times and get the average, plus:

• BerkeleyDB.4.7
• bsddb3-4.7.2 (Python API)
• Linux SuSE 10.1 Pentium 4 Hyperthreads
• Kernel 2.6.16
• Sun Java 5

And here is the code used:

/**
 * Testing for BerkeleyDB
 */
package test;
 
import java.io.File;
import java.util.Date;
 
import com.sleepycat.je.*;
 
/**
 * @author Jorge Alegre
 *
 */
public class BerkeleyDB {
 
	private final static String HOME = "/index/test/";
	private final static long CACHE_SIZE = 128000000;
	private final static Integer MAX = 50000;
 
	/**
	 * @param args
	 */
	public static void main(String[] args) {
		// TODO Auto-generated method stub
 
		EnvironmentConfig envConfig = new EnvironmentConfig();
		envConfig.setAllowCreate(true);
		envConfig.setCacheSize(CACHE_SIZE);
		Environment env = new Environment(new File(HOME), envConfig);
		System.out.println("Data: " + envConfig.getCachePercent() + " " + envConfig.getCacheSize());
		DatabaseConfig dbConfig = new DatabaseConfig();
		dbConfig.setAllowCreate(true);
		long t_1 = new Date().getTime();
		System.out.println(t_1);
		Database db = env.openDatabase(null, "testing", dbConfig);
		DatabaseEntry key = new DatabaseEntry();
		DatabaseEntry data = new DatabaseEntry();
		String sData = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx";
		data.setData(sData.getBytes());
		for (Integer i= new Integer(0); i

Tweet This Post Plurk This Post Digg This Post Ping This Post

Last night I run the Spring IDE for Java. Apart from the fact that Eclipse in my laptops is slow due to low memory, the Spring IDE seems to be heavier than Eclipse. I am used to work with python and Eclipse using pydev, and start-up times are poor, but Spring seems even worse, so like 10 minutes waiting for things to load, etc…

The welcome page looks great, with information from Spring web site, feeds, and download center with cool software like SVN repository, sync of development task with MS Outlook, Trac integration, Grails, Groove, etc… When trying to install those applications got an error, so man, bad luck to me.

Then created a Java project Buscaplus and tested BerkeleyDB. The experience of IDE is pretty nice, getting nice suggestions with the autocomplete.

I still can’t seem to work out the Subversion repository from Google code, will try this weekend. the default repository is CVS, and since the easy-to-install functionality did not work, will have to go the long way.

I run tests for BerkeleyDB and I was not impressed by performance. Aprt from this, Oracle has changed the implementation for Java from the original from Berkeley, so you only have BTREEs, can’t create tables without environments, etc… This was a total blowout, I can’t use it for OctopusDbm, I prefer an original implementation of BerkeleyDB and Tokyo instead of a Java flavour. As part of this, Java implementation does not physically writes data in the xxxx.db file you determine, but creates partitions of about 10MB each. I don’t like this, prefer the original way. If I choose to have a file 500MB with a cache of 128MB then the software should do what I need. So a C++ will have to be worked out for OctopusDbm.

Going back to Spring, I missed an integrated application server coming with the install of the suite. I thought that would come with the installation, so will have to install tmServer based on Tomcat someway.

As a conclusion, Spring IDE is ok, but had problems with me. Many things are integrated like Aspect projects and Spring views and perpectives are cool, but gee, some things did not work and I don’t like that.

[Update Feb 22]

I installed again Spring and everything went perfect. For some reason first install had problems with Equinox. I now see installed the dmServer (Development) and tcServer (Tomcat) as well as the download center works ok.

Tweet This Post Plurk This Post Digg This Post Ping This Post