Eli Weinstock-Herman
"So what do you do?" "I'm lean," "What?", "I'm agile," "What?", "Fine. I make websites."

To Build Automatic Bookmarking – Unsupervised Text Classification

Original post posted on Monday, November 7, 2016 at LessThanDot.com

I’ve been bookmarking all of my online reading for the past 7 years and recently started thinking about using that dataset to dig into trends in my past reading and potentially build a model to start scoring content I haven’t read yet. Even though I have manual keywords for each entry, I decided to look into what I could get with unsupervised text classification techniques to balance out the fact that I had entered those labels over long periods of time.

My first goal is to programmatically extract keywords from consistently formatted blog posts on static website pages.

I learn...

Continue reading

A Communications Hack for Helpful Ideas

Original post posted on Thursday, November 3, 2016 at LessThanDot.com

You’re taking a break from the two week project and bump into Joe at the coffee machine. He’s a little frazzled and starts telling you about this little task he took on Monday that is kicking his butt. While he’s explaining you realize he might have missed a simpler answer, so you ask him if he tried it. He immediately gets angry, of course he tried it, what are you trying to say? And here’s all the ways that wouldn’t work, and so on, then disappears back into his office.

Why is Joe always so annoyed? He needs to work on his communication skills.


Continue reading

Easier Boundary Testing: Keep Parse/Validation/Format rules out of your HTML View

Original post posted on Monday, October 24, 2016 at LessThanDot.com

In a typical single-page application, type and validation logic is entered in the HTML view and we rely on our binding framework or a validation library to layer this behavior onto the form. There are trade-offs to this approach, which are mostly negative as you get into larger, longer-lived applications.

When we embed validation rules and logic into the View, we’re mostly limited to UI Testing to validate them (the most costly layer of the testing pyramid). We also pay an ongoing tax of having to re-define rules every place we put an input field (how do we format percentages for this...

Continue reading

Continuous Javascript Test Execution with WallabyJS

Original post posted on Thursday, October 13, 2016 at LessThanDot.com

After working with NCrunch building and running tests in the background for the last several years, it feels like something is broken when I have to wait for test results or push a button to start running them. JavaScript runners just didn’t feel like they provided the same level of development feedback, whether they were command-line runners with gulp tasks, plugins like Chutzpah, or dedicated runners like Karma.

I’ve posted previously on both NCrunch and Karma, test runners that run .Net and Javascript code continuously behind the scenes as you develop.

WallabyJS is like NCrunch ...

Continue reading

Stop Manually Updating Your Jasmine SpecRunner

Original post posted on Thursday, October 6, 2016 at LessThanDot.com

I’ve used a number of test frameworks and runners over the years, but my first club out of the bag is still running a SpecRunner file in the browser, with all of the dev tools and console output I’m used to from normal debugging sessions. The painful bit has always been manually keeping the SpecRunner file up to date and forgetting every 3rd or 4th file. Having the SpecRunner is valuable, manually context switching to catch it up over and over is not. So let’s automate it away.

In a prior post, I used gulp to keep my static assets up to date while running a self-hosted...

Continue reading

Useful Projects