babs474

Very cool, I love this approach as it is very similar to how my (now defunct) unedditreddit app worked and how a clever app called socialfixer for facebook works. Two thoughts.

1. You are not unblockable. My app was removed from the chrome app store after reddit complained, no review process, no appeals, good luck even communicating with google. Installing an app not from the appstore is almost impossible for a non-savvy user. I suggest looking at a firefox app, their app story is much more democratic.

2. I think this fight is important because a lot of companies/organizations seem to be trying to establish the precedent that they own their website experience end to end, from database to the pixels on your screen. That idea taken to the extreme will lead to computers become opaque devices ala a TV, rather than the hackable playground they are now.

http://socialfixer.com/blog/2013/10/05/facebook-requires-soc...

A single data point is too small to reason about. But lets say there was an ad campaign that was blocked at a high rate, or even just placed poorly where nobody was seeing it or something. Our scheme would assign those impressions a low value, ie it would come out that cranking impressions or lowering impressions isn't effecting conversion rates anywhere else, so why not turn the ad off.

Facebook style ads CAN convert in a tight window, the problem is people are looking in the wrong place for them. Nobody checks out their timeline, sees a coke ad and immediately clicks to buy. Those ads don't "convert".

What does happen is the cumulative effect of those ads put your brand (coke) top of mind. Later when somebody is searching for "snacks" on google and paid search ad comes up, that person is more likely to chose Coke than RC Cola.

You need to look at effect your "brand" advertising is having on your paid search conversion rates. The industry is trying to solve this with "attribution" schemes ie figure out which ads every converted person has seen. (shameless-plug) But my company, Optimine has a much cooler and simpler approach using incremental modeling.

Here is a white paper that I think does a decent job of explaining. http://optimine.com/assets/pdfs/Measuring_Cross_Channel_Valu...

Sounds reasonable to me, especially the memory pressure thing.

It seems like some of the criticism in this thread comes from people who haven't bothered to understand what you've done and summarily dismiss it as stupid. That attitude has been bothering me lately on HN or maybe its programmers in general.

Anyway, thanks for the interesting contribution.

I think the importance a generic coder or programmer will slowly fade away.

Instead you will have people that are a something that uses programming or has the skill of coding. We are kind of seeing this with the data scientist phenomenon.

Instead of a lot of weight being on which programming language you know, Java or C++ the question will be what kind of domain expertise do you have. There will be a proliferation of many DSL type languages to accompany this.

Man there is a lot of negative snark at the top of this thread.

I'm not sure if this system is a good idea or not but I wish some commenters would spend more time comparing their proposed solutions (shared mem, local db, memmap...) to Hammerspace rather than contentless dismissal.

This paper was fascinating when I read it tomorrow. Unfortunately, I had to warn my former future self about their compelling results and prevent myself from posting all those really fun tweets.

I find Jooq to be similiar in some ways.

http://www.jooq.org

Here are things I use for those 3 things that I find really helpful, maybe you will too.

1.http://www.emacswiki.org/emacs/InteractivelyDoThings

2. http://www.emacswiki.org/emacs/DeskTop

3. For 3 I just use built in shell-mode, but what is really neat is saving your shell session to a file "C-x C-w" every so often. Then if you reboot or crash or whatever, emacs-desktop will bring back all your open files and all your shells. I usually have 3, regular bash, a sql session, and maybe an ssh session to somewhere, all with a full record of what I've done, sometimes going back months.

One of the big features of erlang is intelligent hotswapping for building robust telecom systems that never go down, but can be bug fixed/upgraded on the fly. Just because something is not commonly done with current tools doesn't mean its a bad idea, just that it is a bad idea with current tools.

My understanding is that built in java hot swapping is very limited, and indeed would be a bad idea to try hotswapping in production.

If the clojure repl brings us closer to the erlang world isn't that a good thing?

Yes, I think it is very much happening right now. I think that the sophistication will grow and grow as more ad dollars are converted to the digital world.

It is interesting to see the complexity of strategies needed to play adwords increase over time. Big budget advertisers use predictive models and automated bidding to quickly react to changing conditions. I predict some years in the future we'll have crazy high frequency trading robots battling each other like in the stock market.

No I think they are related but different. For instance with pdb or the clojure repl I"ll add entire new functions or classes and try them out.

With gdb or jdb that might be theoretically possible but it certainly is not convenient or commonly done. Typically your stretching things with a complicated one liner expression.

But you are right concepts are pretty close, its just that I find a lot of things that call themselves "debuggers" don't have the ability to add new code. Things that call themselves "repls" dont' have the ability to stop at a point or context like a debugger. pdb being the exception. Why can't we have tools that do both things well.

I always wonder why more languages don't have this, which I call repl at point, or repl with context. Not sure if there is a more standard name. Even clojure with its generally awesome repl support doesn't have it.

I created a library a while ago that does something very similar to what you did, for the groovy language. http://aaronbabcock.blogspot.com/2011/11/groovy-debugging.ht...

It avoided having to pass in variables explicitly by using groovy's AST transform ability (macros) to examine the code for which variables would be in scope and pass those into the repl automatically. I wonder if something similar would be possible in scala?

I dunno, I find the people who don't care about tone and think themselves above considering feelings are the most overly sensitive and easily outraged when criticized.

Its easy to disregard tone when its somebody else's feelings at stake, much harder to look past tone when its your feelings on the line. Nobody is above the fray.

Maybe we need a robustness principle[1] for human communication. Be conservative in tone you send out, liberal in what you accept.

[1]http://en.wikipedia.org/wiki/Robustness_principle

That is my experience too, but I blame it on hibernate which GORM is just a dsl wrapper for.

Its fine when it works, but expect inscrutable session flushing errors when you try and flex it too hard.

Groovy doesn't get enough love as a post java jvm language IMO.

I agree that grails is a bit heavyweight. The people at my former place did some work on using groovy with dropwizard: https://speakerdeck.com/kyleboon/webservices-with-dropwizard... which is seems to be more of a nice collection of libraries, rather than a box you in framework.

mnsure.org was throwing springframework exceptions the day it launched

I understand that strategy. But I would encourage you to instead keep an open mind. You might be surprised at the value you can get from an unexpected direction.

If you have a number of people acting in the way you describe, you end up with an echo chamber, which is what the article is noting.

I guess we will never know how the conversation could have developed since the stories were killed prematurely.

Do you think the comments section of other types of HN stories are held to the same standard? Or do you think the promptness with which these stories are flagged has something to do with their subject matter?

This is really an egregious pattern. Two examples below.

One is an npr story. You can flag these stories all you want, but they still influence how the wider population views programmers and the field. Wouldn't it be interesting to discuss that here?

https://news.ycombinator.com/item?id=6374508 https://news.ycombinator.com/item?id=6360328

The clojure community has great meme: libraries over frameworks.

In my mind java is notorious for having frameworks that box you in. You have to think like the framework or you will have major problems. Think spring/hibernate.

The library over framework idea is that external tools that you use should solve a problem and be composable. You should be able to mix and match easily if you want to throw something away or introduce something new.

Composable libraries can reduce NIH and keep you flexible.

I interpreted it (aurora), as the next gen instarepl. A tool that helps you introspect running code and intermediate results, but more ambitious than just textual substitution of variable values. It would fit right in as a component of light table.

I really liked how some of the inspiration came from how non programmers code. Think excel macros (gasp).

You might have a point about naming, I don't know. In my defense a bunch of reddit tools do the same thing: metareddit.com, uneddit.com, redditenhancementsuite.com

But I'm really curious about what you said regarding sidestepping delete functionality. How do you view search engines indexing and caching of reddit and other sites? Does the Barbara Streisand effect apply? I think that the issue is way more nuanced than no "sidesteppping" allowed.

One of my main motivations for creating unedditreddit was noticing that there was a lot of baloney moderation going on. Cabals of power users can censor quite a bit and it is completely opaque as to who deleted something and for what. I think it could damage the community in the same way digg vote rings used to.

Reddit did not "force" google to remove it, but I know it was at least prompted by reddit. I know this from emails between myself and google.

1. It wasn't disrespectful. The previous version which you refer to directly interacted with reddit servers. I said I would no longer crawl reddit and I do not.

2. There is nothing wrong with charging for something you have built, but you can check out a free version of similar tool here http://uneddit.com/faq.php it also does a good job of raising some points from my line of thinking.

What makes you say that?

Watch out, when I ran into a similar situation with my extension, unedditreddit. Reddit, was able to convince google to simply remove the the extension from the webstore.

Chrome makes it really difficult to install a non app store extension, so your normal user will not do it.

On the other hand, google and facebook have a more competitive relationship so maybe there is less to worry about.

Spot on, about caring about conversions, not the click through rate.

Also true that this seems like a nuclear grade weapon. We have a system at optimine that is similar in a lot of ways, but also a lot less complex. Of course we are representing the advertiser side of the equation so, completely different design goals.

Interesting that HN keeps killing threads on this topic: https://news.ycombinator.com/item?id=6360328

A stubborn refusal to confront the problem is what helps these attitudes fester.