Open but Secure, Opinionated yet Flexible

Open but Secure, Opinionated yet Flexible - securely open, flexibly opinionated - Open but Secure, Opinionated yet Flexible

When we first set up Telicent, we were 100% focused on engineering — a consequence of all the founders being engineers, I think. It’s fair to say we struggled to communicate our open yet secure CORE data platform to investors. Few of them had experienced problems in enterprise data environments, and they ranged from the conservative (“which existing product that I’ve heard of are you a copy of ?”) to the ridiculous (“So what’s the blockchain angle on this then lads ?”). There was so little in the way of common ground. We just couldn’t speak the same language. In the end, it was easier to just show the product to customers who had problems we could solve and grow the business the old-fashioned way. I know, I know, but being a viable business may well be coming back into fashion. As I write this, in the last 2 years we’ve grown the team from 3 founders to a team of 33, and our product is being used in at least three different sectors. No investment funds were harmed in the making of this business.

But, we still struggle to communicate what the platform does, and that’s getting even harder as we start to talk to people in sectors we don’t know so well. In an attempt to rectify this, we’ve decided to spend some money on marketing. Up to this point, every penny the business made was ploughed back into engineering. It’s been an alien experience, but it has forced us to think in depth about our message. The product tagline we’re working with is “open but secure, opinionated yet flexible”. These are two sets of seemingly contradictory features, and we’re now starting to realise this may have been at the heart of our difficulty communicating with people who hadn’t lived the problems we’re trying to solve.

“Open but secure” is probably the easiest one to explain first. By “open” we mean that we won’t lock your data in. We’re huge fans of open data standards, and the Telicent CORE platform is built around those standards. You will always understand your data, and you’ll always be able to move it somewhere else if you choose. However, that data needs to be protected. That means building the platform to zero trust principles, and it also means being able to control access to the data. It’s in the access control part where we’ve really gone to town on the “open but secure” mantra. We implement a very fine-grain access control mechanism that allows you to label all your data (even down to individual fields, relationships, etc.) with security markings that users have to comply with in order to see the data. This is known as attribute-based access control (ABAC) and we apply it at such a fine-grain that you only redact those portions of the data that the user isn’t allowed to see — maximising data availability while protecting the data that is restricted. Secure but open.

It’s probably also worth mentioning that our platform is free, open-source code. We did this so that big enterprise customers are confident they can deal with a startup. The public release is coming in 2024, but our customers already have access to the code. It also means that highly secure enterprises can let their security accreditors look through our code prior to installation. Security through openness.

The “opinionated yet flexible” side of this is a little harder to explain, and it needs a bit of storytelling. Between the three founders, we’ve got getting on for sixty years of experience in enterprise systems, change programmes and “digital transformations”. In these big, expensive programmes we saw the same thing over and over again — two to three years at the start of the programme where architects and engineers did their design work. A bunch of (mostly) bearded middle-aged men arguing over boxes and lines. By the time the programme was ready to actually cut some code, chances are the sponsors had moved on, the technology landscape changed, and the original market/threat conditions were unrecognisable. And yet, all these programmes had a lot in common — data all over the business, in different formats, and poorly protected, that had to be brought into one place (or at least indexed in one place) and made securely available. We saw an opportunity to create a platform built to modern architectural principles that covered all the bases to get you started and could adapt as your thinking matured. We’ve had customers up and running with their own data on the platform on the first day of delivery. We even ship the platform with an ontology built in, because if there’s one thing that enterprises need help with, it’s data modelling.

So that’s the “opinionated” bit. The architecture’s been done for you. We’ve lost count of the number of major programmes we’ve gone into where the team says “You seem to have implemented 60–80% of our requirement”. It’s not a new business model, the ERP systems have been doing it for decades. Where they fall down though, is that they’re expensive and painful to change. We’ve left quite a lot of the CORE architecture open to modification and re-configuration. The whole idea is that you can get started quickly and start to show your users the art of the possible instead of asking them to guess what they’d like (spoilers, that always results in a faster horse). So that’s the “flexible” bit. You can even throw out the ontology and use your own, but we recommend you don’t…at least not until you’re sure you need to.

This whole engagement with marketing folks has been a voyage of self-discovery for us. Hopefully, we’ll get much better at communicating our ideas.