glesperance

It really depends on the job you're trying to accomplish. I'd venture saying that it's way too early for horizontal / massive scale RAG apps.

Most solutions will want to focus on a very specific vertical application where the dataset is much more constrained. That we're this makes more sense.

Also a lot of alpha in data augmentation.

For those like me that were looking for something more substantial on the GraphRag Method -> https://arxiv.org/pdf/2404.16130

Interesting article about long term planing. It does bring forward some interesting thoughts about how one should execute on a plan, but mostly I find it interesting in how this article misses the point of a plan.

I’d like to propose a slightly contrarian, if not, more nuanced view of what is the purpose of a plan.

If one considers a plan as a map that must be followed in minute details, to travel the proverbial territory then, of course, you’re much better with small, unplanned iteration that allow you to learn from each story you deliver.

The image that the author proposes is a very powerful one: yes it is much better to play the lottery by buying one digit at a time, because overall, you’ll cut your losses short in most cases.

However, this only focuses the value of the plan in it’s execution phase, absent from the planner, and the feedback he will get from the divergences from the plan, and the updates he will be able to bring to the mental model / explanation linked to the plan.

Let me be clear: *plans are meant to fail*. Why? The value of the plan is in *HOW* the execution diverges from it, and in so help us to update how we think and reason about the world.

Each plan, has as origin an explanation, theory, of how the world/market/customer works. Changing your explanation of how the world works is the goal of learning. To learn, you must make predictions. That is what the plan is, a simple prediction. Of course we should not expect those to line up with reality 100% of the time. It is quite the opposite: *expect your plan to fail* but most importantly *improve your understanding of the world / your market / your customer based on what you did not predict.*

I'm currently reading Competing Against Luck and it is a truly eye-opening book that highlights and puts the finger on a lot of the luck-based innovation work we've been doing over the years.

I was planning on taking MR Christensen's online class[1] this coming April, but I wonder if this will still be happening given Mr Christensen's passed away.

Would anyone have any good resource, online classes, books, videos, etc that they would recommend to learn more about Jobs to be Done and Disruptive Innovation Strategy?

[1] https://online.hbs.edu/Documents/Syllabus_Disruptive_Strateg...

We've been using k8s-snapshots[1] Very easy to setup and use. Very simple to add snapshots too.

How does this compare?

[1] https://github.com/miracle2k/k8s-snapshots

Have your tried XLM, XRP, LTC, Dash, or pretty much any other cryptocurrency?

They're much faster, and considerably cheaper than Bitcoin to use.

One could presume it's because they want to add XLM ; and want to appear unbiased for now.

Given that bitcoin isn't a medium of transaction anymore and a store of value this makes a lot of sense.

Too bad they aren't replacing this with another cryptocurrency at the moment.

I find this article relevant: https://stratechery.com/2017/cloudfare-and-stormfront-free-s...

That there could have been some covert messaging between Trump's people and an important Russian bank ;

In other words, this seems to be more circumstantial evidence of Russians involvement in the American Presidential campaign.

Really sad to read that this new batch isn't certified gluten free.

Eating gluten free is often a struggle for ppl with celiac disease.

I'm sure a gluten free soylent would be greatly appreciated by the celiac community! (as least I would!)

The way I see it, a gluten free soylent would be my go-to lunch meal when I'm in a hurry or when there are no leftovers.

> 2. the tipping point theory as a whole is a bit dodgy (see below). > (#) Good article on Debunking of tipping point theory - http://www.fastcompany.com/641124/tipping-point-toast

It seems to me that the article doesn't debunk at all the existence of a tipping point in social networks phenomenons.

In fact, it clearly states that it is the law of the fews, "that rare, highly connected people shape the world", that seem to be inexistent in Watts experiments.

If trends are really like forest fires, then there is a tipping point ; it is just not required to have these highly connected people on board to reach it.

This definitely reads like one at least. I can't help but be amazed at the way he uses to publish / distribute it all at the same time using this print spool.

Exactly what I was looking for. Thanks!

IMHO I would expect a big proportion of the sites to break when you browse w/o JS nowadays ; nothing wrong with it, it's just the state of the web now...

I think it would be good to have that on both the client and the server.

Also, as simple as this might be, making it a de-facto standard (on meteor) would be a good thing since it would be a tool that we can count on, all the time. There are too many browsers IMO to make an extension a viable option.

Whitelisting variables would be good as well ; that way we do not get alerted if it is something we expect to be global.

> We've added file-level JavaScript variable scoping. Variables declared with `var` at the outermost level of a JavaScript source file are now private to that file. Remove the `var` to share a value between files.

I think it is convenient to be able to declare global variables like that but perhaps there should be a way to monitor those ; in other words, it would be really convenient to have some form of alert system to notify you when a new global variable is created.

That way, globals created by mistakenly forgetting the 'var' keyword would be easily spotted.

What is the project direction in relation to complex JS apps?

i.e.: How do you guys expect the instarepl idea to be used in apps made of serveral node.js modules/libs or several files meant to be loaded together via RequireJS ?

Here's an interesting list of bugs [1] (some related to path-finding some not) found in SC / Broodwars.

[1] http://wiki.teamliquid.net/starcraft/Bugs

I wonder what part of their path finding algo created a glitch like the drone float[1]

[1] http://www.youtube.com/watch?v=7MN0wjoqmzU

Well, THAT is odd. I'll look into it right now. Thanks for the heads up!

Oh, no, don't worry, you can turn social sharing off permanently. Hit the 'preferences' item in the menu at the top right. In there, you've got all kinds of social features to turn off. And they'll stay off.

Hey Andrew, wavo co-founder here.

We definitely understand that some of our users prefer not to use the social sharing features of our app ; hence we added the capability to turn those off, either temporarily with the use of the "social-sharing" switch at the top of the interface --or-- permanently via the preference dialog that you can find in the top left menu.

There's one too many 'r' in your first url.

It is my understanding that apps can no longer publish automatically at all ;

"In order to provide users with experiences that meet their expectations, we will no longer approve custom actions that publish stories as people consume content. These apps must use the appropriate built-in actions or create a different sharing experience."

Notman house is definitely at the epicentre of the currently booming montreal startup scene.

I really hope they manage to get to match their goal and secure the house for the benefit of all the montreal entrepreneurs.

Hey there, (wavo CTO here)

It is my understanding that running airbrake in the browser does need you to somehow provide the API key to the client[1].

Could you explain a bit more in detail what makes you worry about having the API key on the client?

[1]http://help.airbrake.io/kb/troubleshooting-2/javascript-noti...

You can easily achieve that by redirecting traffic from facebook to the app request page and simply allowing people through when they chose 'cancel'.

You do not get to show the app permission request as a dialog but that te only way you can put a gate before your contet and still allow ppl that select cancel to go through.

Am I the only one to be surprised that so much of this info is made publicly available by the ad sense network?

Moved back from coffee-script because event though all its syntactic sugar is really helpful to program faster, it seems to me that CS lacks clarity and still has too many ambiguities.