Kickstarting Foundation and Goals for 2019!

Would love to hear more about this in detail. Are we looking at a Shopify Partners like ecosystem (just an example which most would understand) where Shopify just sticks with providing/maintaining the core backend and the Partners take care of customisation, design etc work?.

1 Like

Happy new year! Looking forward the partnership program

1 Like

I guess that I am both surprised and moderately disappointed that no mention was made of a strategy for creating and maintaining stable release points of the package. Maybe those of us that care a great deal about this were not included in the 2018 analysis of the success of the project.

Admittedly, there have been some steps to at least reduce the risk of the devastating constant updates that alter the user experience while they try to run their business. I see this in the careful approach to the release of v11 and I think that is a noble first step.

In the opening of this thread, many important key elements have been identified that will also go a long way toward making ERPNext (via Frappe) a success in it’s chosen market. The Partnership program is one such element and I applaud that as a fixed goal.

However, I do not see anything yet that any of my current client base would consider active acknowledgement of their plight and their wishes for the ability to install and even reinstall a stable release that will not be like shifting sands below the foundation of their businesses.

Please have your new leadership at least address their concerns, even if it is to say that you will not alter the current model and will continue to make stability of their systems a moving target.

Many of them have invested heavily in customizing ERPNext in order to overcome some of it’s short comings over the past year. In fact much of the improvements to the v10 POS module came from their investments. However, they are stuck with what they have now and cannot afford to retrain their staff every few months to keep up with the changes. Additionally, they have no way to re-install their current version of software. There are no long term stable release points. So they have had to accept a frozen point in time and pray their VPS service provider does not go out of business because they are unable to re-install their base version of software.

My clients would have been happy with having to pay a consultant to re-install their customizations on a fresh server, but they have no way to get back to a fresh server! Without stable long term release points, they have no way to ever get back to their base version. The playbook is constantly changing and the supporting packages as well. Installing later sub-versions of the same v10 package would require a complete rewrite of many of the customizations. Yet that is their only choice if they need to change their VPS service provider.

Hopefully your new Community Release manager can address in 2019, but I didn’t see anything here to indicate that kind of willingness in the foundation. Hopefully I am wrong, but unless this problem is addressed I cannot see the value in becoming a partner in the partnership program.

BKM

1 Like

Hi @bkm,

I do agree that a LTS would be great, but we also have to keep in mind that maintaining several versions at the same time requires a lot of resources that no-one in this community seem to have/be willing to invest until now (at least I have heard of it but seen no one drop a few $1000 on the table specifically for it).

Anyway, please forgive me to correct you on one important point:

This is not true, since Frappe/ERPNext uses GIT and semantic versioning allowing therefore anyone to choose their version of ERPNext.
A quick search on the forum will guide you through the different steps of this process.
For example:

Of course it will not be as easy as using only the easy install script (there are only two additional steps…), but I do think that if anyone has customizations that can not be easily ported to a new version, they would have the internal knowledge to checkout a specific version of ERPNext and Frappe before migrating a backup to a new server.
And there is always the possibility to fork a specific version if you don’t want to update. Forking will allow you to easily port some important commits merged in a future version (security issues/major bugs/…). In this case you would be maintaining a sort of LTS.

6 Likes

True. But you’ll need knowledge in GIT and you’ll also need to know the frappe terminal commands in order to make this work which could be a major turn-off for the people who want to stick to “Easy Installs” (which you’ve already pointed out).

I kind of understand where @bkm is coming from. As a programmer, I’d like to avoid mending the broken bones of my custom app in order for it to work (only for it to break in the next update). I just hope that in 2019, no doctypes will be renamed. :laughing:

Personally, I’m exited with the ERPNext Partner Program and I’m hoping you can come to the Philippines. We’d love to learn more about how frappe manages to encourage Open Source in India, and maybe, we can do something similar in PH. :slight_smile:

Here’s to a fruitful, exciting and amazing 2019.

7 Likes

What a good initiative it would be good to expand throughout the world …

2 Likes

Thank you everyone for the comments and encouragement of the Partner Program. Will try my best to reply as many queries as I can!

There will be multiple tiers of partner and there will be no partner fee at the entry level.

Interacting with Service Providers over the past few months, we (FrappeTech) got to understand the challenges and opportunities there. Also having observed the way the ERP industry is structured, it has been completely via local partners. For FrappeTech, the goal will be to enable partners by offering trainings, evaluations, services and leads (which will be a mix of both free and paid), and a commission on both the hosted service on ERPNext.com and also our upcoming pricing for self-hosted support.

What FrappeTech brings to the table is expertise, availability and ongoing product investment and the partner will be responsible for local support and implementation. This way, we can grow the ERPNext community much faster. In the current service provider model, there is no incentive for the service provider to work with Frappe and vice versa.

By kind of separating our roles (though these will not be absolute, and cannot be absolute since we are in an open source environment) we will be able to leverage each other’s strengths.

This is not binding either. Service Providers are free to build their own ecosystems, and they will still have access to the same source code and we can still collaborate at the code level.

So the distinction will be that Frappe will run the ERPNext Partner Program (and not be a partner itself) and invest in training and services for the partners. I don’t think the foundation was the right entity for this (and in any project it is not the role of the foundation). The foundation will be a place where we can collaborate on our common objectives, mainly by funding projects that are of community interest.

Your comment about guaranteeing Open Source is that the license of ERPNext is GNU GPL v3. Any company that wants to move away from a pure open source (free software) model will change the licence (examples, Odoo, Redis, Elastic) to a more permissible one.

My personal view is that keeping the software free has nothing to do with the ability to be able to bill for services. If there are certain guarantees that a company that extracts value from ERPNext needs, then they should be willing to pay for it.

Please use your disappointment to do something positive for the community! “Scratch your own itch” is rule #1 of open source. I am sorry its harsh and brutal, but hey! the software that you serve your customers is 100% free and it is a fruit of someone’s labour. Please allow me to express a sense of disappointment to your sense of entitlement.

Yes, very much like Shopify, GSuite, Atlassian and almost every other serious software service!

Version 10 has been stable for quite a while. Maybe client installs should be tested before deploying them. Most good service providers will have their own processes around deploy.

Also the product will be developed at a rapid rate, and we improving the documentation and processes all the time. If you have better ideas, feel free to pitch in!

I think this is a great reason to push contributions to the core that are likely to break from a future release. ERPNext still has a lot of improvements to be built from a polishing and usability point of view.

Yes renames are expensive, but they are also important. For example renaming of “Email Alert” to “Notification” suddenly means that you use it for SMS or Slack notifications. So we cannot rule them out!

8 Likes

Ahh… several of us have read these posts and tried to make them work to no avail. I even paid a core ERPNext developer [name withheld] (a favorite developer of Frappe) to try to create a set of instructions for accomplishing this goal. after several months and many many failed attempts, they, even they are unable to do this.

So… you appear to be speaking without actual experience on this topic. I, on the other hand have been trying for a long time and even spent my own money to pay a core developer to make it work. It doesn’t!!!

The problem appears to be related to getting the right version of “bench” within a major release. For example, one of my clients needs ERPNExt v10.0.14 and Frape Framework 10.0.22 because versions after that have other changes in bench that make their customizations unworkable. Yet trying to install on a new, blank server this version has proved impossible. There are many threads that purport to be able to do this, but in practice, it appears to only work on the last close out version of a major version. It proves impossible to gather all of the specifics needed to replicate a lesser sub-version because too many other pieces to the puzzle have been changing during the point releases and getting them to align like tumblers in a complex lock (at least in version 10 so far) is apparently not possible. If someone is willing to prove differently then I am willing to pay (again) for those valid instructions.

Wow… did you ever get that wrong (and I know you have read other similar posts on mine over the past 2 years). My sense of disappointment is NOT related to my own work. I pay developers to fix anything that I need fixed.

My sense of disappointment is how those at the top of the management structure do not see the how the lack of such things a LTS releases stunts the growth of the project and allows other contenders in the field to potentially overtake this wonderful project with their lesser quality but more stable projects. I know you are familiar with this as you also commented on my “knee of the curve” post about this last year. Your comment this time is just disingenuous. I really do not want to see this project miss opportunities in pushing it’s great ideas forward, but that is really difficult when there is no solid foundation for a small company to base their ERP tool.

Don’t get me wrong here… this misguided sense of not needing LTS versions keeps implementers like myself and the developers that I hire in constant business. However, for this to be a widely accepted package, it needs to be easy for businesses to set this up on their own and be able to maintain it on their own without the need of consultants except for the most custom or difficult of circumstances.

Therein lies my disappointment. We could have a much wider audience by adding stability to the “jumping off point” (ie. LTS ver) where new customers will start their ERPNext journey. Will people like me still be needed? Sure, because many will want to make the package even better after they get over their fear of trying the system.

Good Grief!! Really?!? The question has never been about adequate testing by implementers or users. It has always been about testing something for 60 days and then deciding to use it, but when you then install it on your live production server, it is NOT the same point version you have been testing for the past 60 days and you have no way to get make a fresh install of the point version that you actually did the testing on. THAT is the reason for LTS versions. Otherwise the “testing” you speak of would be a continuous process that never sees an end until you close out the major version number. That is just the reality of the current process.

I agree, but not all contributions (ie. paid customizations) are applicable enough to the general system to be accepted by the core project. I have as many of those out there now as I do accepted contributions, simply because every business is unique and may need something special.

BKM

4 Likes

Dear BKM,

My apologies if I upset you with my comment, I just wanted to point out that saying that something is not possible because you haven’t been able to achieve it so far can send very bad and contradictory signals for new companies thinking of using ERPNext and reading this post.

To illustrate that pinning a specific version of ERPNext and Frappe is possible, I have just created a new installation pinned at the specific versions requested by your client (Frappe v10.0.22/ERPNext v10.0.14).
I have just used the easy install script on a fresh server and some of the commands proposed in the post linked above.

http://pinned.dokos.io

username: administrator
password: password

I also agree that the easy install script may not be usable for quite old versions if there has been any breaking change in the install script. You may also have to adjust the version of the required linux libraries for old releases.
But, again, thanks to Github, it is always possible to go back in time to fetch a specific version of Bench/Frappe/ERPNext and to make a manual installation. That’s the whole point of using GIT and Github.

EDIT: Shutdown the linked instance

4 Likes

@chdecultot

Thank you for the update. Believe it or not, I have spent much $$ with a noble core developer of ERPNext to get a set of instructions to do exactly what you seem to have been able to do in short order.

No offense was taken. I was only working with information from what I believed to be an authoritative and well paid source. I have no problem being humbled by my ignorance when presented with better information.

If you would like to sell me the list of instruction to accomplish this on my own fresh VPS server, then I would gladly consider applying great humility and using paypal to pay for such information. Afterall, I was willing to do it once already and got nothing for my money except to be told it was not really possible.

PM me with how you might like to proceed and supply me with your instructions to do this same thing. I will use those instruction to create a test server and pay my developer crew to apply my customizations. If it all works, then you have enlightened me.

In return, I will take those same instructions (after thorough testing) and post them on the forum under a fresh thread for others in my position to be able to benefit as well.

Easy install script is not necessary as long as the instructions are fairly concise and easy to follow. I frequently contribute to the core and community (whenever possible) and would consider this another valuable contribution.

BKM

5 Likes

Is there any plans for an uplift for the POS? ERPNext lags on this compared to other platforms and given its flexible technology framework, it is easy to make big changes.

1 Like

Hi @bkm,

Thanks for proposing to pay for the creation of the step by step, but I just posted it on a new thread:

I hope it will be useful! and feel free to enhance it with your experiments results!

11 Likes

Wow, that is better than I could have expected. Yes, yes, yes, I would have gladly paid for this and then given away to the forum myself (crediting your work) just so it would be out there.

Thank you for offering it up on your own time. I really appreciate your effort here.

BKM :grinning::+1:

4 Likes

This is great ,
how can one propose or be a partner of erpnext or try to acquire training despite not being from local market ?

i would like to contribute many of my custom apps , but they are largely based on specific customer needs , and thus requires extra effort to make it a pull request worthy only to have it rejected because it doesn’t have need for it in a larger scope .

it would be great if one of the community moderators could open something like a idea thread and pin it globally - where everyone can discuss the ideas they have (ideally should have already implemented on their end) , and discuss if its required for others and at the same time getting enhancement ideas and help and making it pull worthy .

currently it seems people are giving their views on multiple threads and its not synchronous (need to search in multiple threads if you require a idea or people who have same idea as you) ,
so may be a privileged , static for good and global thread can help get everyone on same page . (if its already there , please help guide my ignorant self to it )

thanks for the hard work . :+1:

3 Likes

This is a dilemma for many, specially when you have many custom apps based on specific customer needs but don’t have the energy/bandwidth to make it a contribution-worthy module. Maybe combined efforts of few would make it easy to relay the custom app towards contribution-ready PR.

Mentoring of new developers would help and probably will be part of the post-training efforts within Partnership program.

1 Like

Fully agree with this. I honestly hope that the leadership puts some serious efforts on improvising the POS. Especially on POS receipt printing, barcode printing, customer display, a simplified UI etc. If done correctly, ERPNext can really become a serious contender for stores and small supermarkets especially here in India.

Can someone please consider this and take the lead to improvise the system?. I would love to contribute by doing real world deployment in my supermarket and report bugs/feedback. Honestly tired of using half cooked stuff like Nukkad Shops, Paytm Smart Retail, GoFrugal, Tally etc which are either outdated, or buggy, have poor service and lack customisation options.

Maybe this needs a separate thread to discuss ?.

in such a case you can publish the repositories of the apps here, so at least self-hosters can make use of it.

@HaroldOsprey

We have open source our Offline First POS for ERPNext.

Hope this can help the community.

Cheers,
Chris

1 Like

What a great news!. Now the entire community can improvise the app and probably merge it with ERPNext iOS and Android app?.

Thanks a ton. Now I can confidently deploy ERPNext in our upcoming supermarket.

I was going through the forum and realised that there are two similar projects which concentrate on offline usage of ERPNext (maybe not all features) i.e Frappe Accounting & TailPOS.

It has been a long pending request of the community that we need a strong offline support for POS module on desktop (Linux/Windows/macOS) as the existing browser based implementation has quite a few flaws. Though the TailPOS app seem to address this issue to some extent, finding a proper Android tablet hardware which can take abuse and interact properly with printing devices is not an easy task.

So considering these points can we either add POS module to Frappe Accounting or port TailPOS to desktops in 2019?. With proper support for barcoding, weighing scale, direct printing, payment terminals etc.

I believe this will have a huge demand especially in India as people are tired of using outdated softwares (Tally) or proprietary softwares that need new dedicated Android based terminals (like from Paytm Smart Retail/Nukkad Shops etc which cost a bomb).

If ERPNext Cloud can somehow be made to communicate with an offline client, we can achieve a truly omnichannel retail setup. With ERPNext Offline POS taking care of billing in brick & mortar stores, Cloud which connected to platforms like WooCommerce will take care of rest of the stuff.

2 Likes