History of Talend Open Studio

[Anonymous]

Hi friends! I live in Brazil and I am developing a study about the software Talend Open Studio in my monograph.
I will have a dedicated chapter to TOS, where I will describe its concepts and components.
However, it would like to initiate the chapter with some information about the TOS project.
I want to get the following information:
1) Where year the project TOS was initiated?
2) How appeared the idea to create a Open Source ETL tool?
3) In which city/country the creators of the project live?
Any another information about the history of TOS will be important for my project of monograph.
Thanks !!!
gomes.br@gmail.com

[Anonymous]

Hi Gomes,
1) Where year the project TOS was initiated?
Before the v1.0.0 Open Source (October, 5th, 2006), we performed 3 years on R&D and 1 year on beta test
2) How appeared the idea to create a Open Source ETL tool?
Propriatory ETL like DataStage or Informatica has got a graphical mapping component like Talend tMap. Before Talend, there is no GRAPHICAL Open Source tool that provide this kind of component.
3) In which city/country the creators of the project live?
Our CEO (and cofunder) is in US (CA, Palo Alto) with a Commercial team.
R&D team (with another cofunder) are in Paris (France).
A developper team are in Schentzen (China).

[_AnonymousUser]

ccarbone,
1) You imply that TOS was started in 2002 (2years R&D and 1 year beta test prior to the first 2006 release). I challenge you to prove that this started anytime before 2004.
2) You either misinform or are misinformed when you claim there is no such previous tool. As a matter of fact, the similarities of TOS to another product, especially in its architecture and graphical conventions, are simply too many to be a coincidence. The personal version of that tool was freely available for download.

If you have not been with the company since the beginning, I urge you to check into it. There is this perception that the open source companies are somehow more respectable than others. I think talend should be honest about where they got the inspiration, the design, the architecture, feature by feature design choices, etc.

[Anonymous]

Hi Curious!

1) You imply that TOS was started in 2002 (2years R&D and 1 year beta test prior to the first 2006 release). I challenge you to prove that this started anytime before 2004.

Before 2005, the project was completely "closed" (not available on any open source repository). Versions prior to 1.0 were based on totally different technologies (C++...).
In 2005, we decided to rewrite a big part of our code (obviously, we capitalized on our previous development (architecture, template system...)).
And we decided to open our source code when we felt ready for it (ie clean code, functional, well documented, good performance...).

2) You either misinform or are misinformed when you claim there is no such previous tool.

You extract this from its context. Ccarbone's post says: Some proprietary solution offer such a graphical mapping component (with closed code by definition) but as open source tool, talend's tMap component is unrivalled (for now!).

As a matter of fact, the similarities of TOS to another product, especially in its architecture and graphical conventions, are simply too many to be a coincidence. The personal version of that tool was freely available for download.

I'd be very interested in knowing what tool your are thinking of (if not included in the list below).

If you have not been with the company since the beginning, I urge you to check into it.

ccarbone is in the company from the begining (even before the company has been incoporated).

There is this perception that the open source companies are somehow more respectable than others.

I think talend should be honest about where they got the inspiration, the design, the architecture, feature by feature design choices, etc.

There's no point reinventing the wheel. OpenOffice Writer/Calc got inspiration from MS Word/Excel and that's for all users' benefit, don't you think that's what matters in the end?
We all get inspired from each other...
Really, we never tried to hide the fact that our inspiration comes from other products' parts:
including:
- Informatica ( www.Informatica.com)
- Oracle Data Integrator ( www.Oracle.com)
- Genio ( www.Opentext.com)
- BODI ( www.BusinessObjects.com)
- IBM Datastage ( www.IBM.com)
- AB Initio ( www.ABInitio.com)
and, possibly many more!
By the way, if you look in detail into these products, you will find A LOT of similarities!! BUT if you find any code line taken from their code, I'd be really surprised 🙂
Regards,
Fabrice
Talend's COO

[_AnonymousUser]

Thank you for acknowledging the 2005 start date. The question still remains: can you prove that the start date, in any shape or form, was earlier than mid 2004? Even if it was not clean code. I suspect that your claimed 4 year total dev cycle is really 1.5 years or so.

Many of your examples you list are not valid as they too are recent efforts. For software professionals, it is quaint to suggest that the similarities between two products must be checked line by line at the code level. Here is a question: where did you get the idea to distinguish main rows vs. lookup rows and which of the examples you cite has it?

[_AnonymousUser]

DataStage has main rows/lookup rows. Pentaho Data Integration has them... Looking at the components list qua functionality TOS has been borrowing a lot from Pentaho lately Where there's a big divergence now in functionality is that TOS is splitting up per database type, while Pentaho/Kettle use 1 component for all (most) databases. I'm still thinking on which kind of access is better.
I don't mind that tools borrow from each other, I guess it's normal... for users all the better. All of the tools want to have the most complete feature set else users will complain the other tool has the functionality. In open source you see the gradual changes, in closed source you only see the result every year or every x years.
I do find it funny that TOS pretends to be the biggest and most downloaded open source ETL lately. PDI/Kettle was open source (with graphical interface) latest december 2005 (you can really start to compare in version control who was really first first) and from scratch to version Kettle 2.4 it already had 100.000+ downloads : http://www.ibridge.be/?p=24

[Anonymous]

Thank you for acknowledging the 2005 start date...

Thank you for reading again my answer. I don't think you got it right. Regarding the start date, you are free to believe me or not, but whatever the proof I could provide you with, this will never be enough for you, obviously.

Many of your examples you list are not valid as they too are recent efforts. For software professionals, it is quaint to suggest that the similarities between two products must be checked line by line at the code level. Here is a question: where did you get the idea to distinguish main rows vs. lookup rows and which of the examples you cite has it?

Once again, read my post more carefully. I provided you with the list of products that we got inspired from. I didn't see the point of your questions but I answered them with goodwill, why don't you answer mine, now?
Regards,
Fabrice

[Anonymous]

version Kettle 2.4 it already had 100.000+ downloads

Well, great for them! I really mean it. We are glad that Matt Casters and Kettle (PDI) meet such a success. We never considered them as our competitors. Actually we met Matt at MySqlCon in Santa Clara, a couple of months ago. He's a great guy, very smart. We are very impressed by his work. We even started talking about a possible collaboration on some stuff.
Regards,
Fabrice

[_AnonymousUser]

There may be 2 "curious" posters... maybe more ... so wait until one of the other guys wants to answer your questions.... probably the other one meant DataStage, that was the first impressions I had when I saw tmap 😉
As in my previous mail personally I don't really mind that the tools copy from each other... there are only so many ways in which to do a thing. There's no pointing in doing something in a different way if a good way exists.
I've been playing with both TOS and Kettle. There's room enough for both I think on the market, some people will have a preference to one or the other. But both I think have the problem of being open source and that it takes a kind of momentum to start them both off.
I did a couple of evaluations for projects and in 90% of the cases we still end up with OWB or another closed source tool. Companies are "scared" to use open source or dual licensing, the bigger the company the more fear they seem to have.
P.S. I don't like the splitting of types of databases, but that's just a personal opinion.

[_AnonymousUser]

Fabrice,
I understand that you are the COO and in the management role but I must say you sound defensive.
I had a very simple question at the beginning and you answered by confirming that 2005 was the start date but I am afraid you did not answer the original question. Even if it was completely "closed" and based on different technologies before then, there must be a way to confirm it. Absent that, the claim that it took 3 years R&D and 1 year of beta testing to get to 2006 simply does not hold water. At best, it is misleading to your clients. (While on that topic, Mr. Carbone is listed as having joined the company in 2006, but you claim he was there even before inception. Again, a little consistency is needed.)

I honestly don't know what your question to me was. In any case, I find it most productive to close one issue before splitting it into ten new ones. And if you cannot answer kindly, I'll happily consider the matter closed 🙂