πŸ”

Launch of multi-device sync + Memex goes 'back-to-beta'

With the update to 3.0.0. Memex we'll launch our brand new sync and backup architecture, offering users multi-device sync, auto-backup and tablet support, also change in licenses.

Also Memex will go back-to-beta with all ongoing subscriptions waived.

What's changing:

  1. Memex now syncs across unlimited devices and provides auto-backup to the cloud.
  2. Memex Go, our mobile app, now supports annotating websites on tablet sized mobile devices
  3. Effective immediately, we are cancelling all ongoing subscriptions and Memex can be used 100% free to use while Memex is in beta.
    • You can support us with the Pioneer Plan that gives you 3 years of Memex subscription at a 60% discount, once we reinstate the subscription plans (Q1/2, 2022)
  4. Memex and Memex Go are still local first tools. To facilitate the backup, sync and collaboration functionality, we also have a Firebase backend.
  5. β€£
    Read Privacy Note

    So far the data stored on your device and on Firebase is unencrypted, as we do not have the capacity to implement strong privacy protection infrastructure yet. Rest assured that we will not look into this data or exploit it otherwise. What you don't share with people is meant for your eyes only. We know everyone can say that, but we took strong measures to prevent organisational exploitation of your data. For example, we don't take VC money and our investors have a capped return because we want to remove the incentives for profit maximisation, one of the root causes of the need to exploit data or attention. We want to stay build tools that people are willing to pay for.

  6. Changing our open-source licenses to the 'Source Available' license 'PolyForm Noncommercial License 1.0.0'., allowing free use and modification of the code unless used for building a commercial product out of it.
  7. β€£
    More about our decision for this step (click triangle)

    This applies to all upcoming code (3.0.0+). We're also, for now, close sourcing our cloud, sync and collaboration infrastructure. (Note that all existing code stays open source)

    We have to pick our battles well. We are already making our lives much more difficult by not taking VC money and instead using the Steward Ownership investment model. To protect us from being co-opted through profit maximisation incentives, our company can never be sold and has a capped return for team/investors. However, this requires us to be profitable. We want to prevent that our business model and application can be copied 1to1 and thus have years of work jump-starting a better (VC) funded competitor before we had a chance to stand on solid feet.

    This scenario probably wouldn't help to have Memex long-term available, improved and maintained. There would also be one service less with the ethical values that Steward Ownership nurtures because it disincentivizes profit and growth maximisation and through that give users more data openness, more interoperability, more social responsibility, more data ownership.

    Of course, this is all the ideal state, and we are still unsure how this plays out. This step as a more pre-emptive measure to prevent the above scenario, until we figured out how to exactly structure this so Memex can be as open as possible while enabling it's ability to stay a sustainable business. If you want to join the conversation on figuring out what could be a good way of making the software the most open source it can be, you can add your thoughts to our forum section about it.

    Next steps on making Memex more open again:co

    The easiest step for now is just to close source some critical components and prevent commercial use of the rest, and work our way to classic open source licenses from there.

    So far we see ourselves going through this in several stages.

    1. We closed source our core library (memex-common), which contain the multi-device syncing infrastructure, and are keeping Memex in a source-available license. That memex-common is completely closed source right now is only temporary logistical issue and will for the time being unfortunately prevent people from building the extension in their forks.
    2. We are separating memex-common into a closed source and source-available part, so that anyone can still build different Memex versions for local-only use still and the cloud infrastructure components stay closed source . Working on this is more of an issue of available resources and time than general will to do it.
    3. We are releasing useful libraries in a full open source licence again. Note that all code up to this point was already open source, so you can get inspired there if you need it.
    4. We figure out together what's needed to make the Memex extension again fully open source and have it work with the cloud. For that to work we still have to figure out some technical and strategic details. We feel it would be for example awesome if Memex could just be adapted by anyone and still hook into the cloud/sync/sharing backend.

Why are we waiving your subscriptions?

With the launch of the multi-device sync, a substantial part of our infrastructure is new, our collaboration features are in an early state and generally Memex has a lot of rough edges. At this point we don't feel comfortable charging you for the Memex service.

Instead we want to work with you in the next 6 months to get Memex to a state where we feel like the price matches the value you'll get out of it.

What you need to do

  • If you're using Memex already: Just follow the instructions on the dashboard on how to migrate.
  • If you are an existing premium subscriber: Your remaining subscription time is allocated as credits to your account. You'll be able to use them as soon as we'll charge subscriptions again.

What took us so long to build this sync?!

Mostly, the work of the last 1.5 years was pivoting and cleaning up of bad product decisions (and the resulting idealistic technological decisions) of the past. The web history full-text search functionality, our core feature since the beginning, turned out to not actually solve a frequent and painful enough problem. Because of the sensitivity of web history data, it required us to build a much more privacy focused product (e.g. local first, e2e encrypted sync) than we had the funds to build well. In short: it was a big distraction.

In hindsight, when we go tour first big grant (about 150k) we should have paused, and really deeply think about what we and our users really need from this product. Instead we let the narrative around the full-text search, and the grant requirements for decentralised products cloud our judgement.

In the past 18 month, we put a lot of effort into changing the way we research, prioritise and spec out our work and started from scratch with our user research. As a result, we were able to make more revenue in the past 2 months than in the previous 8 combined.

Quo vadis? Where is this going?

From the start, the goal of Memex was to build a tool that helps to lower the impact of misinformation and social polarisation on our decision-making.

To get there we want Memex to become is the most low-friction way to privately and collaboratively curate, annotate and discuss content online.

We all do so much research or reading online, filter for quality content, write comments, take notes and interact in different ways with the content we are exposed to.

All this information would be immensely useful signals for our peers to get to quality content quicker or understand the nuances of the topics at hand.

The problem is that right now it's really hard to share or receive those signals, collaborate in curating quality content or discuss the important bits of content we come across. It requires tons of copy-pasting stuff around into Google Docs or other note taking applications, where we constantly produce single-use information.

It's also not an alternative to post every good content we read to Twitter. It would be relatively low friction but would quickly spam our time lines. Sharing a condensed blog post or tweet storm with all highlights can take hours and does not really help peers that are not proactively following this work.

On the other hand, if you want to get those signals from your peers the only way is to doom scroll or time line or ask them individually about their opinion, so it again takes too much time.

Also, often we are still in research mode and not ready (or never will be) to synthesise our thoughts in a Tweet or blog post. A quick recommendation or highlights that others can see while they are reading the article would already be very helpful and low friction to share though.

But the fact that I read it as a good read would already be super helpful to others.

Sometimes it would help to just know that 5 of my friends read an article I am about to read, and see what important sections they found, even without any commentary.

What we need is the ability to share recommendations, highlights and notes on articles with 1 click and have conversations around an article while we read it.

Next Steps on our roadmap

In the past 4 months we have done almost 100 user onboarding and research calls and have a clear idea on what you folks need. You can read a full

about our findings here, although since writing this some things have changed.

Up next are:

  • Local PDF support Open PDFs from your local file system and annotate them privately or collaboratively with your peers.
  • Collaborating in different context Often, you read the same article but want to make notes for different groups, like work, friends, the public. Right now, all notes that are shared, are shared across all collections the page is part of. With this update you'll be able to selectively choose to put notes into different collections.
  • Improved Editor The current editor is really really basic. It supports some markdown but does not at all have a good UX of structuring your text. We'll implement tiptap.dev's editor. This will also give us some future adaptability to provide e.g. referencing blocks saved in Memex or a Roam/Obsidian integration for PageRef suggestions.
  • Saving and annotating images So often you want to save or take notes about images in an article, but you cannot right now.

Get up to 58% discount on our upcoming subscriptions

Memex is going back-to-beta. Reason is that with the new sync and the early collaboration workflows, a substantial part of our infrastructure will be in an early state .

We’re not feeling comfortable asking for a subscription fee before this is rock solid. We want to work with our supporters and users to make a product that is worth paying for.

In the mean time, you can support Memex with the Pioneer plan, giving you an up to 58% early bird discount, and us the ability to operate with speed and without taking VC money.

More info

  • Our subscriptions will launch in ~Q2, 2022 with plans ranging from $5-20.
  • For your $150 one-time support you'll get $360 in subscription credits and at least 3 years subscription time, even if you pick the most expensive plan (min 58% discount)
  • For your $10 per month support you'll get $15 in subscription credits each month (33% discount)
  • Do note this early bird offering is intended as financial support during our beta phase. Subscriptions will likely range from $5-$20/month once they launch and what exactly they will cover is still under debate.