Odoo: The new OpenERP | Hacker News Reader

Odoo: The new OpenERP

odoo.com

44 pointsseyz44 comments

If you look around there is no "beautiful ERP" solution out there and the status quo is "you can make a very capable ERP or a "pretty" ERP but not both". I understand the the difficulty of striking that balance but I think by applying the right kind of modularity to the app design (sort of like openERP but even more granular and forgiving) you can arrive at a solution which invites businesses of all-sizes but can grow with the company into very complex processes.

There is HUGE opportunity in this market and I believe one big reason is that a new generation of userbase will expect and demand a better UX than any current offering.

In fact, I suggest a great startup idea would be a to take the openerp codebase and wrap a new UI/UX around it starting with one vertical ("niche") market and working out from there. (i.e. "A beautiful ERP for medium-sized businesses focusing on your growing warehouse, sales and purchasing needs built on the solid foundation of openERP."). Not saying it would be easy, but just worthwhile. I have a feeling the Odoo rebranding points to an effort for a UI overhaul which acknowledges the opportunity of a great UX.

> built on the solid foundation of openERP.

Solid foundation is a joke. OpenERP is probably one of the shittiest pieces of code every written in Python. Take a look at their ORM code. And, of course they still use floating points for accounting [1].

[1] https://github.com/odoo/odoo/blob/master/addons/account/acco...

Disclaimer: I work on the Tryton project

By curiosity I went to your website http://www.tryton.org/

It doesn't even have a screenshot, and the main description does not tell clearly what it is. The only apparent link beside the menu bar is "Validate HTML". There is no link on the items such as "Accounting" in the description.

SAP certainly has pieces of a "beautiful ERP" with Fiori. https://www.youtube.com/watch?v=mYbn9Q6P0cU

This is nonsense. An ERP has to run your business and operations. It has to be extremely reliable. Users couldn't care less about it's UX, in fact, the standard ERP solution, SAP, looks terrible, but it is also quite ok to use. In a big company an ERP error/fault can generate millions worth of damage, are you still thinking a prettier UI is a great start up idea for an ERP?

Finally, you forget that IT departments in big companies are just a tad above trained monkeys and your software solution has to take that into account, in fact, it's the first piece of your service: it has to be easy to maintain. It's probably here where you want to focus on to compete...

> This is nonsense. An ERP has to run your business and operations. It has to be extremely reliable. Users couldn't care less about it's UX [...]

Indeed. When a person's livelihood depends on using your software, they have a pretty strong incentive to learn how to use it effectively, and put up with all its imperfections.

I'm sure some employees actually revel in how backwards and difficult to use their business software is. If everything were automated, effortless, and required no expertise, they'd be out of a job.

> Finally, you forget that IT departments in big companies are just a tad above trained monkeys and your software solution has to take that into account

This generalization is the equivalent of "Every SF startup engineer is a high schooler who happens to know some javascript".

Think before painting with a broad brush, especially when its that department that might have to give signoff for your product.

There's are diff between erp and mrp.Much focus on material resource planning. I do property erp software and customize for government.It really really complicate with all of layer of procedure . Taking other code base without experience can be painful and complicated. If any people here dealing with complicated procedurelayer.please share your experience.

By the way, the core and base modules (and in theory, all community modules, since they're derived works) of Odoo are Free Software (APGL) and available on Github: https://github.com/odoo/odoo

Disclaimer: I work for a company that implements solutions based on Odoo for small and large companies. So yeah, it's possible to make a living working on FOSS, if anyone still doubted it :)

they still use their custom-own-built orm ?

Looks like it: https://github.com/odoo/odoo/blob/master/openerp/osv/orm.py

Yes, throwing it out would be suicide, since it'd break every module out there. They're revamping it while keeping backward compatibility, though, and the new API is definitively nicer.

http://www.slideshare.net/openobject/presentation-of-the-new...

I cannot understand why anyone would not use SQLAlchemy if they're building a database-backed Python application.

True today, but not 10 years ago :)

Plus Odoo ORM and SQLAlchemy have different purposes.

I'd love to chat with you about my open source ERP, Stockor (stockor.org) Email is nathan @ that domain.

I'm currently firming up the plugin interface for it and am really want to get it right so that third-parties can easily build on top of it.

OpenERP could've really benefited from some usability expertise, and if this site is any indication, Odoo may suffer the same fate. If you're going to put this much time/money into a new brand, how could you miss so badly?

In my experience with OpenERP, the end users always took a back seat to the developers. Granted, ERP software is inherently complicated, but with each new version you could tell they made an attempt at a simplified interface and never hit the mark.

>if this site is any indication, Odoo may suffer the same fate.

If it weren't for the "formerly OpenERP" next to the new name, I wouldn't even know what the product was supposed to be. Looks like an app store of some sort for tablets.

Spare a sentence or two of explanation, please? Maybe delete one of those rows of giant icons if space is tight.

OpenERP is still leagues above the million dollar paid apps of Oracle and SAP. I'm not saying you are wrong. Just how bad this particular segment of software is. It is wripe for disruption.

ERP software is huge, simply because there so many different ways businesses works, all the regulations you have to comply with etc etc

It would be very hard for a startup to build it from scratch.

Not sure the rebrand is what will attract new customers to OpenERP.

A couple months ago, I evaluated ERPs for our business. OpenERP was one of the main contenders but ultimately the manufacturing module failed to meet our requirements and custom development would have been too costly.

I was also shocked to get a glimpse at OpenERP's internal corporate structure. Despite the open source nature of the project, their business model puts them on track to become yet another big, slow, expensive ERP company pushing massive service fees.

Unfortunately there's just no great solution in the ERP world. Requirements for this category are so complex that by the time a system becomes adequately feature rich, improvements in software technology make it feel obsolete. I wonder if someone will crack that nut.

Exactly. Every company's business process has it's own uniqueness, so if you try to evolve it into a mass-market tool (as opposed to designing for it up-front), you end up with a Frankenstein system that is enormous and complicated. Then your company starts morphing to match what you're selling.

If some YC'ers want to make some serious cash, go build the JIRA of ERP. Make the system: pluggable, VERY configurable, give it a sexy UI and don't require an army of consultants to implement. Want some reasons? Pretty much all your potential customers hate their existing ERP system. Your competition is staffed by sleazy salespeople that sell nothing but snakeoil. Your market is filled with companies of all sizes who will literally throw cash at you to make them more efficient.

What are the alternatives to OpenERP today ? (I mean, free and open sources ones, with acceptable UI)

Even the paid ones have horrible UIs. In fact i've never been lucky enough to see an MRP / MRP2 / ERP system with a usable UI :-(

Pointers to nice ones greatly appreciated since it's useful knowledge...

The phrase I've never heard once in my IT career is "I love my ERP."

Why do you think that is?

I think the amount of work that goes into them, regularly I hear 12-18 months, and that they aren't usually engineered towards intuitive processes or interfaces. I imagine features rather than experience is a driver.

There is http://tryton.org/

however I don't know what you mean by "acceptable" UI

For me, an acceptable UI in an ERP means this ERP can be used to do basic tasks by someone not having any PhD ni ERP Usage :)

The homepage doesn't do a good job of convincing any business why they should adopt it. At the moment, it's too technical for businesses to care.

Compiere is another FOSS ERP, but it's made in Java

ERPNext - https://erpnext.com

GPL v3 - Python, Javascript, MariaDB

Github - https://github.com/frappe/erpnext

(I am one of the developers)

Less features than openERP and not open-source but more usable http://www.myerp.com

disclaimer: I'm a developer @myerp

I'll take the opportunity to plug my attempt, Stockor @ http://stockor.org/

It's built in Ruby and currently uses ExtJS for the UI. I'm currently rebuilding the front-end in Backbonejs.

Would you mind talking about your reasons for switching from ExtJS to Backbone?

Sure.

The biggest driver is that I want it to work equally well on mobile, including phone sized screens.

The other is the file size and load speed. Right now the build is pushing almost 3mb, and takes around 10-20 seconds to load. That's kind of ok, since the workspace is intended to be a very long running app that folks will use all day long, but it still really sucks. You can view a demo of the ExtJS workspace at: http://stockor.com/demo

Plus it's just kind of ugly in my opinion. It looks like a desktop app that's attempting to be delivered in a web browser, which fair enough it is. I'd like it to be something more stylish though.

As for why I'm picking Backbone vs Ember or Angular, that's a bit more complicated. The main reason is that I'm very familiar with Backbone and have used it for years (I'm a contributor to DocumentCloud).

When I made the decision to rewrite for mobile, I did do a pretty in-depth framework comparison and Backbone still comes out ahead in my opinion. It's light enough that it's super easy to understand and figure out why things aren't working the way you like, while giving you just enough structure to keep the app on the correct path. I am going to use a view/model binding lib though, but haven't settled on which one as of yet. Probably backbone.stickit.

Email me at the address in my profile if you'd like to chat more about it, glad to answer any questions you might have.

Icons!

As a designer, I've probably never seen so many icons on a page.

I've considered working with OpenERP and looking at becoming a consultant in it (ERP is $$$ and very useful), but last I checked the OpenERP site (2? years ago?), the demo site was half broken. Glad to see it's gotten a refresh and a genuinely web 2.0 name. ;)

Apparently the devs have churned it into a SaaS offering, which is also very cool.

I installed OpenERP some days ago, just to see what it had become. I found it nice. Even though I think you would still need to be really documented on how to use the different modules.

Well, my company sell training packages at affordable prices :D

is OpenERP contractor fees that large?

I know SAP and Oracle Fees are high, simply because its quite hard to get opportunity to work on it, and all the certifications. So there aren't that many people around.

Well, I'm not sure what the rate is. But I know that it supplies a very valuable service, and other things in the space are very lucrative. I imagine that part of the lucrativity is the value that it provides to businesses.

Salary does't always mean value provided.

Salary = Supply/Demand.

I wouldn't say it's hard to get an opportunity to work on it. The demand is high (which is why the fees are high).

Being good at it is hard, the certifications are sort of meaningless. You need to know a lot of different things, and it's very challenging to have the comprehensive toolset needed.

They also discovered how to succeed as a tech company: Better use at least three 'O's in your brand name.

"Choosing the right name is hard. To find the perfect name, we analysed the names of top internet companies. We discovered a direct correlation between the company valuation and the number of 'O' in its name. The graph below shows the average valuation of the top 10 internet companies according to the number of 'O' in their names.

With three 'O' in its name, 'Odoo' is in phase with our ambitions."

Reply on news.ycombinator.com