Project vs product (and some comments to product itself)

As we like to make our statements using various communication channels/media it becomes pretty hard to respond to anything.
Yes, guilty of that too. So not claiming any superiority here. And every medium has it own public. And everybody has it own preference

However makes it also hard to find 'the right' place.  Anyhow more or less responding to https://twitter.com/Sweet5hark/status/1303065894533427201

In my perceptions it about the perspective. From internal perspective (so within the organization of TDF) LibreOffice is viewed, managed, and functioning as a project.
There is a community. There are volunteers at QA, translation, marketing, support (ask) and number of volunteering developers.
How the developers at the eco-system partners are prioritizing bugs and refactors is outside my sight. However I assume most of the bugs are picked by the developers themselves.
So the got quite a lot of freedom related to the work on LibreOffice. So also a project approach. Not sure if there is also more commercial product breach with dictated projects; strict deadlines and such.

LibreOffice for the end-user is pretty often associated with a product.  What do I mean with product. There quite a lot of dimensions to unwind here.
* From marketing perspective. What are you selling. In the SaaS model customers don't get a product, but more service. However they software is still the product the company is selling. And there are maybe still binaries to be installed.
* A product in end-user perspective. The product is what you paid fore. If it tooth paste of license to a certain software. With SaaS this get sore complicated. It's product (say Office Suite) is wrapped into service.
Except you aren't the end owner anymore. That part of buying a product and owning it (or the license to use it for ever) is stripped out.
* The results of a hobby (software) project; they application itself it the product of the coding effort.

They end-user uses LibreOffice as a tool to get his job done. They use LibreOffice as product to get certain things done. Linguistically he can't even 'use LibreOffice as a project'; it simply sounds weird.
So the result of a successful (community/hobby) project is a certain product. And this product is also seen as product by others. So LibreOffice being a project is an internal view; but not necessarily for the outside world.
It's a project for those whole are involved. And it managed, organized in a (hobby) project manner. So no commercial drive to get certain things fixed or the need to for new version with new features (to sell)

And even this isn't black/white. We are release a new version every 6 months, with release notes promoting new features. Somewhat odd for an organization not being product driven.
So an organization who doesn't need to squeeze out something new every 6 months or year (to get people to pay again for the upgrade)
So there is no binary distinction between Project/Product. However it's true that LibreOffice is less product focused in commercial terms.
Which means bug stay open longer which commercially might be a no go. Introduced or revamped features don't always match expectations (as the are releases somewhat unfinished).
And the product/feature owner (the developer who build it) might have moved on to something else

The issue I see with going 'commercial' with paid a Product (they software that say quality expectations of customers going up. I paid, so I might assume X/Y working properly.
So with paying customers that responsibility is taken for bugs. And those get solved within a certain time frame. And regressions being less tolerated (pressure at QA and DEV).

You can extend the Product (software) with services. Special features, special support for very specific bugs, consultancy etc. However this is an extension of the main Product (Software).
They main product (Software) needs to be in good shape. I don't see customers paying for extended services (derived products) to solve self-inflected regressions in the main Product.
Even if they main Product (software) is being shipped free. Extended services are simply addition to core product no a replacement.
Say A has to use extended service to repair a regressions caused a change made by developer Z to solve an issue of paying customer (B). This must be calculated into they price of B or in the price of the core Product.

And for the record. A product stays a product even being free. If I get beer for free from the brewery, it's still get product (beer). Even a free product can sometimes not meet the expected quality standards.
Say brewery gives a way beer for free but it's spoiled. You might not go back and request something better (because it was for free). However doesn't go well for the brand reputation.
The brewery could also give creates of beer away for free. And extend his Product with insurance that each pipe of beer will taste well. You have to pay for that of course. And if not OK you have to fill a form. Bring it back to certain location.
Waiting for approval. Ultimately it will be resolved. Do you want to take the risk when hosting a party? Or do you prefer to pay upfront with a quality guarantee in advance (where normally out of the question of the taste will be OK).

So what should the brewery do; go for better product or giving beer away for free with insurance? And what would the customer do. Would he prefer free beer with uncertainty if it taste well? Possibly ruining his party.. Or go for save.
Again this whole topic is fluid. So you Product could be pretty cheap; with some warranties about quality. And some extended services. Which is actually the case. Something called Product Warranty.
The insurance premium (for broken product)  is calculated withing the price of the core product.  So minimum quality warranty; if not return you get something new. Or in software terms.. we fix it ASAP (and for free).

Regards,
Telesto

Hi Telesto, all,

Telesto wrote:

From internal perspective (so within the organization of TDF)
LibreOffice is viewed, managed, and functioning as a project.

and

LibreOffice for the end-user is pretty often associated with a
product.

True, that is a good description of the status quo.

I cannot speak for Bjoern, but what resonated well with me is the
"project over product" (or also called "community over code"
elsewhere) line.

Which says, to be successful in the long term, the advise is to
generally favour community needs over user needs (and as a corollary,
also focus marketing on attracting more contributors, rather than more
users).

And I agree with that approach. Of course, a sweet spot & a virtuous
circle is when your community is motivated to build a great product;
and in turn growing numbers of users of said product can be won as
contributors. The discussion was not about that desirable state
though, but for the case that there is a conflict.

Cheers,

-- Thorsten

Hi,

Telesto wrote:
> From internal perspective (so within the organization of TDF)
> LibreOffice is viewed, managed, and functioning as a project.
>
and

> LibreOffice for the end-user is pretty often associated with a
> product.
>
True, that is a good description of the status quo.

Quite. A notable distinction is that the first reflects mostly a reality, while
the second is only a perception that has no base in reality. Therefore one
shouldnt build strategy on or conclusions from the latter, and focus on the former.

And yes, the latter might be a hugely _popular_ perception in a specific
audience. That still doesnt make it one bit true.

Telestro wrote:

If I get beer for free from the brewery ...

Software and beer markets are not acting the same. If I give away one beer for
free, that leaves still a lot of beer of the same quality to be sold even in
the same town. If I give away one specific software for free, nobody will buy
software of the same quality anywhere on the world anymore.

So lets not do beer analogies, please.

Thorsten wrote:

I cannot speak for Bjoern, but what resonated well with me is the
"project over product" (or also called "community over code"
elsewhere) line.

Which says, to be successful in the long term, the advise is to
generally favour community needs over user needs (and as a corollary,
also focus marketing on attracting more contributors, rather than more
users).

That is a very good summary of the guidance I tried to express, thanks!

Best,

Bjoern

Hi,

Telesto wrote:

From internal perspective (so within the organization of TDF)
LibreOffice is viewed, managed, and functioning as a project.

and

LibreOffice for the end-user is pretty often associated with a
product.

True, that is a good description of the status quo.

Quite. A notable distinction is that the first reflects mostly a reality, while
the second is only a perception that has no base in reality. Therefore one
shouldnt build strategy on or conclusions from the latter, and focus on the former.

And how does this differ from the development teams behind MSO, OnlyOffice?
Those are projects for the people involved too (say developers). They are attached to their project too.
Where the developers are proud off in general (there will be more internal struggle)
However for marketing/ company boss of the same company the 'product' of the project
will be the product for the company to be sold to the public.

And where is difference with TDF/LibreOffice? Even the eco-system partners are selling a product.
LTS version of LibreOffice? They accents are different..
And somehow the eco-system partners are alive and kicking. Money tree? Found a pot of gold?
Sponsor with very deep pockets?

And yes, the latter might be a hugely _popular_ perception in a specific
audience. That still doesnt make it one bit true.

True and False are binary. I think it's more how you're looking at things.
So both are true in their own way. It depends on how you present it.

Telestro wrote:

If I get beer for free from the brewery ...

Software and beer markets are not acting the same. If I give away one beer for
free, that leaves still a lot of beer of the same quality to be sold even in
the same town. If I give away one specific software for free, nobody will buy
software of the same quality anywhere on the world anymore.

So lets not do beer analogies, please.

Point taken :-). Software isn't a consumable, but a capital good.
However following you're line of argument.. is LibreOffice in such a bad shape?
Based on your assessment I would expect the user base to be larger ;-).

So there must be something preventing that the prediction didn't come true?
Software quality/feature completeness/Marketing?

Software decays; must be maintained (and updated) the keep up with rest of the
hardware/software eco-system. Else it will get unused at some point. Even if free.

LibreOffice 5.3 and down doesn't run on Ubuntu 20.04 (by default); Harfbuzz error
There is a currently a 'work around', but it's a start of a process kicking in.

Also the world might change and require different things (Cloud version) or support for large images.
40 megapixel images at home computer 30 year ago? Likely?

But yes, if are shipping outstanding, fully support, fully featured software with regular bug fixes for free,
it's likely that everybody would use that.

But this would require someone to bring in the capital to make it free for the rest of the world.
Hardware, Infrastructure aren't free. Software licenses/MacOS are not free either and so on.

Point was more about the whole idea to 'ship LibreOffice for free' (without strings attached).
and only having an extended support as a product for the eco-system partners.

I'm personally don't see this flying. I think the product must necessarily be selling the 'Software' and 'Services"
But feel free to proof me wrong.

The compromise would be 'fresh' being free. And stable (LTS) being paid. How they paid variant would be for retail customers. you can argue about that.
I proposed already something about pay you're price. So prefilled 5 euro (my adjust by income of the country); which can be adjusted by the customer.
So we get also an impression what people are willing to pay (for better support/more bug fixes etc).

Only rudimentary notion :P. Different directions are possible of course.

Thorsten wrote:

I cannot speak for Bjoern, but what resonated well with me is the
"project over product" (or also called "community over code"
elsewhere) line.

Which says, to be successful in the long term, the advise is to
generally favour community needs over user needs (and as a corollary,
also focus marketing on attracting more contributors, rather than more
users).

Traction.. more users; more publicity; more potential contributors?
And they development (code) is mostly done by commercial party's.
I don't see this changing soon. So revenue question kicks in.
Where LibreOffice more obviously becomes a product.

And one thing I wanted to point out that this will affect they project/community.
Attitude of the user changed (becoming customer) and attitude of developer changes (focusing on end user experience
with different priority's. And less 'open' to enthusiasts or 'random' contributions.
In a project you're allowed to break things and keep it that way for a while, this bad for reputation for commercial product.

Best,

Hi,

And how does this differ from the development teams behind MSO, OnlyOffice?

If you cant see how setting up a project where contributors work according to
their own agenda, motivation, goals and ambitions compared to being directed by
a centralized management which is not even contributing themselves is very
different, I dont know how to help you. The first is the promise and hope of
multi-agent open source.

I proposed already something about pay you're price. So prefilled 5 euro (my
adjust by income of the country); which can be adjusted by the customer. So
we get also an impression what people are willing to pay (for better
support/more bug fixes etc).

Sure it would be nice if users would donate more. However, even if donations to
TDF would increase by a factor of ten it would barely be enough to sustain the
status quo from that alone without ecosystem and unpayed volunteer contributions.
That is, if TDF would even be suited to effectively spend that money. The
progress on tenders this year continues to show that to be challenging.

So donations alone are unlikely to have an significant impact on the projects
health now or in the near future.

is LibreOffice in such a bad shape? Based on your assessment I would expect
the user base to be larger ;-).

You for some reason seem to still measure the success of LibreOffice by the
size of its userbase. The LibreOffice project and TDF are the wrong place to
discuss that, here it is all about contributors. _If_ you really want to
eagerly talk about the size of a user base, I suggest you apply for a job in
marketing at one of the ecosystem companies. Because unlike here, that is where
that is where the number of users are relevant.

In a project you're allowed to break things and keep it that way for a
while, this bad for reputation for commercial product.

LibreOffice is no commercial product and it has to take care about motivating
its contributors. Case in point, that was one of the reason I created bibisect
for LibreOffice QA back in the days. It was not primarily for users, but to
keep the balance between different active contributors: With bibisect, each
contributor was owning up responsibility for the impact of their contributions
and made those transparent to the rest. But the forum that decided what is
acceptable and what is not are the active contributors -- and nobody else.

Best,

Bjoern