ベン・ギルド (Ben Guild)

This project launched a bit too early, honestly… perhaps when it was only two-thirds or so finished. I promoted it only amongst a few friends and with a casual blog post, as well. There were some features that were never turned on, although its core functionality was finished and stable. — Ultimately, though, I ended up changing employers just after its first public release, and decided to shut the platform down shortly thereafter.

“Shh…” was a sort of IRL social network. First, you'd connect with a few core friends via a trusted medium, such as in real life either by scanning QR codes generated in-app or by exchanging links through an established chat thread on some other, non-anonymized messaging platform. Then, you would simply post messages either selectively to friends or to everyone you knew in the app, and they would only be able to guess who they were from. Messages could then spread throughout the entire network, not just to your directly connected friends.

Even though you were directly connected to your friends, they wouldn't know if a message someone had written was from you, or from one of your friends, or from one of their other friends, or from one of anyone's friend's friends, or beyond. Messages that you had read could spread to your friends, then their friends, and then their friend's friends, and ultimately not stop forwarding until “dying out” at the compounding mercy of each recipient's voluntary or involuntary reactions and some server/client-side algorithms. The message forwarding flow in some ways resembled behaviors in Conway's Game of Life.

Furthermore, the system was designed in a way so that all messages were encrypted both locally on device and also end-to-end in transmission. This meant that the data traveling between friends and various other friends' devices (and beyond) was not readable by anyone “in the middle” who was not a designated recipient… such as the app's server and cloud system providers, anyone's internet service provider, any engineers working on the app's platform, nor even if all of the data were to leak and be shared publicly somewhere online. Decrypting these “256-bit encrypted blobs” en masse would be a cryptographic undertaking of considerable size.

I'd written the iOS app's entire client-side codebase in Objective-C, a language that I'd also begun more so to distance myself from in the years leading up to this app's retirement. My major, personal iOS projects until then had all been written in this language, but my interests had shifted increasingly and more dramatically toward Swift with its newer versions, ABI stability, and ultimately with the unveiling of “SwiftUI.”

It's tremendously challenging to launch social networking products, but they are fun. Communication is the backbone of the smartphone's popularity, and some chat applications are incredibly successful. I would love to have a simple, anonymized feed to post messages to with friends (and beyond), but developing products takes a lot of time.

“Shh…” was discontinued in early 2020.

Mobile17 was a mobile-content tool enabling the simultaneous creation and editing of personalized cellphone ‘ringtones’ using any original full song found on the user's computer as a music-source. (via iTunes, MP3s, iPod, etc.) Editing of songs into ringtones took place primarily via its website and was assisted by crowd-sourced statistics/voting of which parts of the songs sound best. Ringtones could also be created offline on smart-phones using its iPhone and Android apps, and data was transmitted back to the website via a centralized API. In late 2011, a native Windows Phone application was co-developed with Microsoft to enable its users to sync ringtones from the Mobile17 website directly to their handset, without any complex downloading or additional installation.

The site either hosted a file for download, or delivered content directly to the user's handset upon creation. The company was acquired in 2012.

While finishing school, I started a project called “Café Wifi” after coworking out of cafés for a summer in New York City with friends. The app was designed to make it easy to attribute and populate detailed data about the best places to work from while on-the-go… anywhere in the world.

Although coding began in 2015, I didn't go “all-in” on the project until 2016, and towards the end of that year we were accepted into a VC/accelerator program in Tokyo offered by Digital Garage. — By the end of the project, we had enabled users to swap seats and meet up at the best walk-in workspaces worldwide, and had essentially built-out a first version of “Uber for seats” at cafés. The seats were provided by other users and/or café staff at establishments, either for free or for a small amount of money paid via Apple Pay.

On the back-end, we triangulated locations and tracked statistics of Wi-Fi networks as published by the mobile apps, and displayed aggregate data both on the web and in-app for each venue.

One of the coolest features that we had was the ability to do multi-level/multi-area indoor attribution of venues. In other words, you could tag and configure attributes (such as power outlets, seating types, smoking/non-smoking, outdoor seating, food availability, cryptocurrency acceptance, and more) in various areas of each known café and coworking space, not just for the entire space itself. All of this data was synced automatically in the background between various devices and our central database as edits were made, and would even resume or recover at a later date in the case of spotty connectivity.

Café Wifi was discontinued in March 2018.

Accounts was a simple network for sharing your accounts and usernames that you use for web services selectively with groups of known contacts. The goal for the service was for people to keep in touch using whatever apps from the App Store as they always have, but then to use Accounts as a consistent, reliable, and centralized index of all of the people that they know. Accounts was not only a powerful address book that updated itself, but was designed to keep people in touch and manage their contacts for the longterm.

Users would receive notifications when someone updated their information, or if there were new apps installed on their devices that could be added to their profile. They would choose who to share their basic contact information and particular accounts with, and wouldn't feel obliged to post any updates or worry about their privacy given that there was no actual user-generated content being shared within the app!

Accounts was discontinued in April 2015, with plans in hand for a successor product.

The player had a single “life” with each play, or up to two lives by unlocking the second “undersea bonus level” with an in-app purchase of $0.99 USD. The game was otherwise free to play, but if the player crashed on the initial “outer space” level after having achieved a certain score/points threshold (from shooting and avoiding flying objects and platforms), they would reach a cutscene of traveling to Earth over the Atlantic Ocean, and either be prompted to pay to continue the game or instantly transported into the second level thereafter if they had already chosen to make that optional purchase. The player could in theory continue playing endlessly for free on the first level by not crashing into platforms and by successfully destroying flying objects, even as the game increased its speed, but the player would then never reach the bonus level or cutscene.

RocketBro was available on the App Store and free to download and play through the end of 2015.

Took from concept to reality in just three days.

Accomplished recommendations by mathematically analyzing your listening habits and comparing them with online sales rankings of a selection of artists to find missing or previously unenjoyed (but similar) tracks.

The project's website was released to extremely positive reviews that same week. It cost virtually nothing to run by relying on Amazon Web Services and other cloud offerings for scaling and stability.

Bitsmash did not aid in distribution, but instead gave a centralized database of statistics that identified content to likely tangible product matches (such as actual Books or DVDs found for sale on Amazon.com) and could also geographically pinpoint (via Google Maps) where users were spreading content to or from. The site offered Billboard-style charts showing the demand for certain types of media, as well. Closed this project in 2009 to refocus and apply knowledge gained to both Dubspace and Mobile17.

Modified and tuned Volkswagens with friends for years, and simultaneously launched Dubspace.

Although the site received great praise from its users, its market was too saturated for organic growth. I later applied its technology and design attributes to Mobile17.

App Map tracked live app usage activity, anonymously, worldwide amongst its users.

Users who had downloaded App Map participated in an anonymized app-usage sharing platform that was great for app discovery. Their true identities were never tied to their locations, and social connections could only see each other's app activity when opted-in.

Created as a working concept for Digg, a socially-driven news and bookmarking site launched in late 2004. Digg had become one of the internet's most popular websites. It organized news into tech-oriented categories with the stories ranked by their popularity, but lacked personalization features for users.

Accomplished building this by measuring users' interests in keywords and topics from their past bookmarking history. The site was an instant hit and actively discussed in the blogosphere.

Bitoogle (later renamed to “Yotoshi”, before being sold) received a lot of traffic and turned into a profitable (through online advertising) side-project very quickly. — It was eventually purchased from me by an entrepreneur based out of Tokyo, Japan.

Being a teenager at the time (with limited financial resources), I built servers for the site out of older computers and parts that I had around the house. One of the better machines was actually assembled in a Dr. Martin's brand shoebox after I ran out of cases.

Wrote this Windows software in 2003 to enable users on a computer network to search and exchange files in a serverless, distributed peer-to-peer network. Each computer connected to one-another and created an instant, encrypted network for sharing of files, without the need for a centralized or “always-on” server.