Should We Look for the Killer App for Solid?

This post is in reaction to Looking for the killer app for Solid.


There were some, admittedly, very interesting ideas mentioned in this thread so far. But I’d argue that none of them so far could be said to constitute “The Killer App for Solid”. Now, why do I have this feeling? :thinking:

Let’s elaborate on that…

Back to basics

I have been confused about and discussed the positioning of Solid in the past. It was advertised on the community website. Since then the landing page of the Solid project has markedly improved. Now, Solid in itself constitutes of a rather low-level set of standards. What they offer - when implemented - in terms of Unique Selling Points are all Non-Functional Requirements.

Listing from the landing page:

  1. Data ownership
  2. Interoperability
  3. Open data
  4. Privacy & Security

Then the landing page mentions what Solid can be used for when widely adopted:

“Data stored in Solid Pods can power ecosystems of interoperable applications where individuals are free to use their data seamlessly across different applications and services.”

Note that I have highlighted “can power” because it refers to the Vision of Solid…

Vision

The vision is about offering the NFR’s above, then combined with ambition:

Universal, global adoption of the Solid suite of open technology standards.

In other words, Create a Better Web.

Killer app?

My original confusion, and also what tempered my enthusiasm to delve deeper, was that Solid in her positioning overloaded the terms of “App” and “Application”. Seeing them as the ‘old way’ of doing things, and offered “Solid Apps” as the new and exciting alternative. Solid essentially made a call-to-action to completely reinvent the web, and fully rebuild from the ground up. Every app should be replaced with a solid app along the way.

No!

Imho this is unrealistic and will never happen… at least not in the way this community thus far approaches achieving the Vision.

Going back to what started this thread:

The answer is obviously that all these application domains could profit from Solid. In fact, any business domain could. And where they do, they start forming aforementioned ecosystems together.

Though there’s absolutely nothing wrong with building your own Solid-enabled apps, I feel that in general the original proposition is what has led this community astray in her pursuit of making Solid a resounding success. And here I highlighted “community”, because I think that e.g. Inrupt is not led similarly off-track… they are right on top of the money!

Let me explain more, why I think that is the case…

Who’s the customer?

Take these two statements:

  • Software has eaten the world. There are billions of software projects, and trillions of dollars invested.
  • Solid is not an end-user technology. The USP’s it offers are mostly not manifesting as end-user features.

Do you think an app will be the disruptor here? An InstaSolid, or SolidFace, SolidTube maybe? Of course not.

Solid not being end-user technology, means that end-users are not the direct audience that should be targeted. They are the developers, the product managers & marketers, the IT decision makers, any techie involved with software development!

Solid target audience is anyone involved in software development

I see most people in this community dedicated to Solid’s success trying to target end-users with things like “Here’s my MVP of AppXYZ with Solid in it! How do you like that?”. Most end-users will not be easily impressed, I think.

And even the majority of developers looking at your example apps won’t be. In a world where new projects pop up every second, they’ll give a quick look at the code and go “Meh, NodeJS + Typescript. I am Elixir and Go… next project”.

Note that Inrupt et al doesn’t have this positioning. If they enact their vision, they’ll be a unicorn before long.

Stripe   – – –   — — —   – – –

It is funny that when I duck out “stripe” the first pages of search results all refer to the unicorn company of that same name. It is well chosen :smile:

Let’s look at the definition of the original word:

Stripe

(strīp), taken from Idioms:

  • (one’s) (true) stripes
    • One’s true or honest beliefs, thoughts, convictions, biases, desires, etc.; one’s real personality, character, or disposition.
  • earn (one’s) spurs
    • To prove one’s skill in a particular area.
  • earn (one’s) stripes
    • To prove that one is deserving of a particular position or designation.

Appropriate, right? Rock-solid name for the Stripe.com company. And guess what… Solid has equally nice idioms :smiley:

I am, of course, mentioning this because Stripe is doing things the way that the Solid community should be doing things - again, imho - to be successful and attain the Vision. What does Stripe do then, and do so well?

They do not provide an app, or apps. Stripe has productized a concept, namely that of “Payments” in any way, shape or form. And their audience are not end-users either. It is software developers across the entire spectrum, anyone involved in IT. In this they have proven their skills in a particular area:

To make it dead easy to integrate Stripe whenever one thinks Payments.

Solid … A Killer Technology !

This then is the gist of this long post…

Forget the pursuit of the Solid Killer App !

If you manage to create a killer app, it will probably not be because it is a Solid app (though Solid support may well be an important contributing factor). Your app will be killing because you built unique end-user features in a particular existing or new domain.

Instead strive to position Solid as a killer technology. Right now it is mostly an emerging standard specification. It is a set of draft papers. (Though by all means you should continue to build new apps, if that has your passion)

Killer technology means:

  • Tools
  • Libraries
  • Documentation
  • Processes
  • Best-practices
  • Events
  • Education
  • Advocacy

These types of things. And if there is a library, and it is NodeJS + Typescript, then the community should actively seek for “Who implements this in Golang, who in Elixir, and who in Java / Kotlin?”. For documentation it means “Who translates to Spanish, who to French?”. For processes “What are the current procedures? How can we make that easier? Can a tool suppor this”. If something is end-user facing e.g. a UX component library “How can we make this as frictionless as possible?”

Most of all what this community really needs is more active members.

Members coming from all across the target audience. Here advocacy, education, events come firmly into the picture.

Community of Action

This community of technologists - it has been discussed before - is not well-supported by those developing the standards. They have Inrupt and some other parties as central focus and a commercial audience as target customers. That’s fine, and a good business decision, from their perspective. Unicorn status, fame… possible, if executed well.

What will their products look like? There’ll surely be a lot of open source. I guess the business model behind this will be mostly open core technology and value-added services. How long will it take for free software, FLOSS, to catch up? Dunno.

Right now, by developing end-user apps, most of this community is effectively in a “wait-and-see” modus operandi. Passive, rubbing hands to see what next great things become available. It is a pity, and it may be a missed opportunity later on.

AFAICS there’s no one taking community initiative, doing the community chores and trying to make this tech lift off and skyrocket in the free software movement.

Right now you are in a Community of Interest and for Solid’s Open Vision this should become a true Community of Action.

Question is…

Do you want to take charge?


Battery emoji from Apple

10 Likes

Yes, indeed. Note that I am not saying that Inrupt, SolidProject are doing anything wrong. On the contrary. Also in general I wholly subscribe to the cause and purpose that Solid stands for, and addresses with their USP’s. And I greatly admire the fact that - in a world where walled gardens are all too common - there is passionate work to craft new and open standards. I am even not suggesting that this community is not aligned with Inrupt. After all, that is for the community to decide.

What my long post boils down to - let’s say the TL;DR of it - are consideration of two questions that involve the community itself:

  1. Killer App vs. Killer Technology?
  2. Community of Interest vs. Community of Action?

Note, btw, that these same questions also apply to the Fediverse in a way. The W3C Recommendations have seen great success so far, by early adopters embracing them and creating great apps. Much more apps, in various domains, are being developed. But increasingly the interoperability between those, becomes an issue, and ever more complex (other than on a case by case, app-by-app, basis). While app developers focus on their own projects, they need to see that there are win-wins to be actively engaged with SocialHub where standards are evolving. There’s need for coordinated action, and v2.0 versions of the specs.

At SocialHub I am addressing this in: Organizing for SocialHub Community Empowerment - Fediverse Futures - SocialHub

Both Fediverse and Solid are fragile gems. Their future is by no means assured. Great disruptions may come, e.g. in the form of the EU Digital Services Act or even Twitter’s Bluesky project. At the time when Semantic Web was at the top of the hype cycle, the entire dev community was excited by it. Yet it failed, and now remains as another of the many tombstones in the landscape of the Decentralized Web (and I know it still exists in some circles, and - who knows - may be revived in the future).

Strong communities create strong ecosystems.

3 Likes

Great analysis. I have challenged many developers from MIT and Academia over the last 5 years to show me how to solidize my medical questionnaire app for patient controlled sharing of their medical history at each of the yearly 100 billion encounters worldwide. My conclusion is that there is no business model, the available technology lacks basics; yet, the idea of personal control of data is brilliant. In medicine it saves time and lives. In medicine trillions of $ are spent on useless and visionless projects. Plenty of money should be available for this Solid-demo project. I see the point made and agree with the analysis that the basics fail to make this project a success. However, the proof is in the eating of the pudding. Create a sample-project to demonstrate how powerful Solid could be. How to Solidize Medical Dossier-Tool for patients

5 Likes

The statement “Forget the pursuit of the Solid Killer App” is frightening to me. I think killer apps need to be our vision in the abstract. One killer app emerging out of nowhere isn’t realistic, but without an active app ecosystem we will have very poor adoption. With an active app ecosystem I think we can hope for killer apps. I think Ruben’s blog says it best (Designing a Linked Data developer experience | Ruben Verborgh), we need to be focusing more on apps.

To me this mirrors the software development debate of agile vs waterfall. If we build an awesome platform, and then throw it over the wall for adoption, we will crash into the wall instead. If in fact the “Solid target audience is anyone involved in software development”, then we need to be continuously integrating with app developers. Diatum for example chose not use solid for their project (An Agile Approach to Self-Sovereign Identity)

Well, I meant that in the context of the Solid Killer App thread’s topic. In general it is always worthwhile to be in a quest to find the Killer App, but do that on its own merits. “Find the Killer App for Solid” is similar to “Find the Killer App for XML”. You are not really talking apps here anymore (devtools, yes), but trying to find the answer to “What makes this a killer technology” that leads to its widespread adoption.

Also I am not saying that building example apps is not a worthwhile effort. But if the whole community were involved with that, who is left to make the technology itself more accessible and attract newcomers from all across the dev/IT community? The already near overworked core team and Inrupt et al, I gather. This is the wait-and-see approach…

If the vision of Solid is adoption across the board, then this is the audience. Adoption is what needs to be facilitated, the adopters will help do the integration and the community will grow with that, become stronger a force in the tech landscape.

With pure end-user app development it would be more like “we rewrite the web into the Solid image, in hopes people (software development) become so enamoured with our apps, that they dive into our JS codebase and distill things for their own app”.

Seems a harder proposition to me, to really take on. But both roads have value to be followed and they can reinforce each other.

One killer app emerging out of nowhere isn’t realistic

Multi User Domain :wink:

The project is essentially a platform to build Multi-User-Domains, where very generally the “Domain” part being provided by RDF storage and some specifications/protocols for reading/writing that storage, we can build a client which compiles a sort of game on-the-fly from a federation of those Domains

Then the Multi-User part is Solid. Storing user data on a Pod which they can take with them between worlds is what gives it the capacity to be “infinite”, like Doctor Who jumping into their Tardis

We started it in January 2021 though and mostly for fun/curiosity, I’m joking about it being “the Killer Solid App” that we might need

What I’m saying here is that regarding the chicken-and-egg problem implicit in the value of Solid “when its everywhere”, as a developer Solid the Killer Technology is what made me want to be an egg - now - before it’s everywhere

I’m not saying that the RPG data stored on a Pod is going to unlock Solid’s chicken-and-egg problem on its own, but I do think that having examples of interesting Solid apps which use the tool now is a great way to expose developers to their own ideas, result in more apps, until eventually we cross the threshold and there’s enough of an ecosystem to make the long-term objectives immediately accessible

So in essence I can see how the abundance of “Killer Apps” mark the defeat of the chicken-and-egg problem, but I think that “the Killer Technology” is ultimately what will convince someone to write a Solid app to reach that point. I think the apps we are looking for now are best if they’re not waiting on the promise of “Solid everywhere”, but they build things which help us get there

I think that the subject of the political approach to building a framework is related to the Killer Technology question, since those parties are seeing value in the technology because of its political implications, although doubtless that is tied to the promise of “Solid everywhere”