drdoom

You know this is a special pet peeve of mine. I have stopped myself several times from complaining about this very

For a period of a few years, a few years ago, I started shopping very frequently for pants, shirts, etc. I once caught myself wondering why that was the case. The answer surprised me: I was not happy with the stuff I was buying. They would not fit comfortably after the initial wear.

Then I noticed what was really the problem:

1) Pants are getting shorter, mostly in the inseams - the distance from the belt loops to bottom of the zippers. (In other words, pants now "sit below the hip", there is no more those that "sit at the hip").

2) Shirts are getting shorter too. They don't go as far down under your belt as they used too.

So, just these two combinations leave a good part of my lower belly exposed quickly and easily, something I do "not" like.

3) It is quite hard to find pants with basic options, like pleats and cuffs. Everything is flat fronted, it seems.

So, it is easy to see why no one is shopping any more unless they really have to.

Just my 2 cents :-)

Here is an insider point of view of the reason Tim Cook missed his sales goals as an Apple engineer describes it:

https://www.youtube.com/watch?v=cXuVAi2bHdA

Congratulations on seeing the opportunity and quickly moving to do something about it. It is unique enough at first sight that you got early coverage in the press, which is very helpful.

Quick question: I did not see anything unique to "old geek" in the website, other than the URL of course. I guess it is an implicit assumption by both job seekers as well as job posters.

On that note, where would this concept be headed if other job sites added a simple attribute called age (or something similar but more palatable) where job posters could specify their preferred age range, and job seekers could search on it?

Just curious:

What kind of a demand is there for a pipes-kind of product or even a customizable/searchable rss/feed integrator?

How much would a typical user be willing to pay for it?

A few days ago, in a moment of 6-degrees of youtube, I came across a flat earth video. Out of curisoity to see what their beef was, I watched one. It was interesting. One of the claims was that all of the earth images we have are not actual photos or pictures per se, but rather, artists renditions (I think they used a different term which I don't recall at the moment). Apparently, this point was admitted to by NASA too. And they had pictures of earth from space (taken and distributed by NASA) over the years where continent sizes change drastically on the same circle. One of them even had "sex" spelled in the clouds.

Watching this, it too looks like an artist rendition. Notice how the moon moves in a straight line. And while the earth rotates around, the moon doesn't. That is fine as we see only the same side/face of the moon but remember: this is taken from far far away so, at that distance and angle, you should be able to pick up a change in the moon's rotation as well. Here, we don't.

Interesting indeed.

OK. But I am curious how it maps a phone number to a carrier. My understanding is that carriers do not publish the list of their customer phone numbers.

It seems like textbelt does not know this information (or at least when you sign up or send a text message). Does it try in sequence to send the text to all carriers one by one until one succeeds? If so, I'd imagine textbelt would get blacklisted pretty quickly.

It would be interesting to see how the mapping is done.

Most of my Prolog experience was in academics - for my own research as well as for maintaining some existing packages. But I did get to use it outside as well. There are many cases where a controlled language is needed or comes in handy and solves a lot of issues. This is where Prolog shines - plus you get to use a lot of the previous stuff you developed as well.

When it comes to an "AI" language vs. typical language with an "AI" library, I will just repeat a phrase that is often used in the field: what you thinnk of as an AI problem today will no longer be considered AI tomorrow, since it will then be well-understood and implemented in many places. People used to say that in response to the (seeming lack of) progress AI had made to date. A good example of that is Siri or Amazon Echo. Not many people consider it an AI problem anymore, or so it seems.

I have used Lisp, Scheme, Prolog (and multiple dialects of each) for extended periods in large projects. Each was a pleasure to work with. They all have their own advantages. I am not sure if there is value in arguing over x feature of one language vs. y of another. As a matter of fact, it was fun to write Prolog interpreters in Lisp, and Lisp interpreters in Prolog, for example. (Many courses / textbooks (used to) present these as side projects to do and the task is simpler than it sounds.) One could say Tcl is really an extension of Lisp or Prolog in its core concepts of syntax, data and program equality, style of interpretation, etc.

I agree wholeheartedly with the last paragraph of Mr. Ousterhout's reply here - and I must say, a smart, classy and almost Tcl-ish way of a jab at its distractors:

http://vanderburg.org/old_pages/Tcl/war/0009.html

I came across Tcl/Tk while doing Motif in a large C-based project. I simply couldn't believe how powerful, simple and succint Tk was compared to Motif (or anything else since). And it ran on Windows, Unix, Linux and Mac OS, to boot. It was too late for that project to switch, but all my other projects have used Tk if they ever needed a UI.

Similarly with Tcl. I still think the best introduction is Mr. Ousterhout's original Tcl/Tk book. It stands as one of the best language books on my shelf. Combined with Tk, one can put together a working application prototype in no time.

Of course, this was eons ago. Nowadays, Tcl offers one of the best environments with a complete set of packages ranging from web servers to image and sound processing. Plus, you can distirbute your program and all of its media and support files easily as well.

I believe Tcl has been mischaracterized and has suffered in terms of open popularity. But for insiders, it remains as one of those secret indispensable Ninja tools that is used over and over again for competitive advantage.

Hello,

I haven't read the book but I am planning to.

I caught the last few minutes of your interview on NPR the other day. You said you built the FB Exchange. I made a mental note to look that up later and now serendipitously, it shows up here. Anyhow, not familiar with it but it sounded like some sort of an interface to Amazon. Would you care to elaborate?

This.

I have given similar demos. You first explain what you are going to do. Before the demo, they go all go into "no, it is impossible. I have been in this business for 20 years. Blah blah." You demo it, and once in a while, they get it, their jaws drop and they appreciate it. But mostly, it is a suspicious disbelief, followed by adversarial challenges where they insist you disclose everything under the covers right then and there.

I have even heard this: If this was possible, why hasn't IBM done it?

Sometimes, it is pretty depressing.

There is no other dependency in the encoder/decoder functions so there is no point of failure there. Those functions were developed a long time ago and have not changed. They work with any stream of data and automatically chop off the input after the first few hundred characters. This contains more than enough info for the developers to see exactly what caused the error.

In any case, even if, as you say, if the crypto handling system failed, and it returned nothing, there was no critical information that was lost. Only information on an error message, which you will hear from the user anyway.

We have been doing this in all of our web applications for over a decade now. Essentially, whenever there is an error, we don't just display the error message as-is. All the technical or trace info it contains seems to scare users. So, we simply encrypt it and display a base64-encoded version of it. It also gets saved into a log file.

Users are more comfortable with this way: they simply copy/paste the text to us and we have all the info we need.

> My view, which I've held for many decades now, is that glib and frivolous invocations of Hitler, or Nazis, or the Holocaust, are a kind of forgetting.

So true. I have myself observed / felt this several times. My experience is that the facts of any discussion are laid out pretty quickly at the beginning. There is something that prolongs the discussion and heats up the arguments on all sides and I believe it is our collective emotional baggage, or lack of it. Some twist their words to mean anything just to stay in the conversation, some dig in their heels in the hopes of never having to be proven wrong, some stick to the bare facts but ignore others' emotional investment, etc. And some just enjoy trolling as if it is a spectator sport where they flame both sides but without any meaningful contribution of their own towards a resolution.

Sadly, I have not found an approach that works better than to quit the discussion cold turkey.

> When she was at Google, Mayer oversaw the approval of each daily homepage “doodle.”

I love this new career path. Beats working years just to advance to manage your old team.

Now I can see the logic behind all the attention on emojis at the Google I/O event. They will be our new leaders.

Some people caught on earlier to the direction she was going (i.e., down). It was so alarmingly clear that some journalists even floated the idea that she may in fact be an agent of a competitor to bring Yahoo down. In other words, she was suspected of being an insider agent working for the other side.

Good points. I agree completely.

Yahoo has also gone from a genuinely useful and relevant aggregator to a full-blown click-bait shop.

Perhaps this is how she thought Google makes its billions and she wanted to emulate it.

Sad but she will be richer for it.

New emojis !!! :-)

I am looking forward to when they release the version where they are generated on the spot with their new GPU improvements.

Let's not go to what I think they should do.

But for so much resources and combined talent, I would expect more than new sets of emojis. They way they were dishing it out was as if they brought peace on earth. Honestly, I would have felt embarrassed to mention it even in passing. (And what was with all those jeans and dress code in general?)

Also, if you are copying Apple and Amazon, which are known for their attention to detail, how about improving things a bit? At the end of the voice demo, the app purchases the tickets (without showing prices, locations, times, etc. (and how do I set up those in the first place?)). Now, the kicker: the phone displays a QR code and asks the owner to show it at the ticket counter at the movies. Himmm.... Really? How about skipping that too and just buzzing/vibrating/texting the user with that info when actually at the movie theater? After all, they spent a whole lot of time talking about context awareness, location awareness, NLP processing, etc. Didn't Siri do this much years ago?

Anyway, too many other things like that. I don't have time to list them all. I am just saying that a lot of people expected something more, and more cohesive.

I didn't know about Google voice search, but I am betting it was just a voice interface to a normal search. Nothing like answering questions, keeping contexts, etc. like they were advertising today. Siri, however, started out like that from the beginning.

What? They can't wait to send it over later when you actually do use it?

Oh, no, Google Deflate has started...

I watched the Goggle I/O event for 10 minutes. First, they talked about how they copied Siri.

And now they are talking about how they copied Amazon Echo.

Unlimited resources. Wasted...

Is there a way to get in touch with you?

What is with cb.vu? It takes you to the shell prompt...

That sucks. There was a time when being entrepreneurial was added to job descriptions as a nice to have. I went through something similar a long time ago; the reasoning was just as specious.

Good luck.

Sorry to hear that. Do you remember what question or problem they asked you?

But you realize that before they mock it, they first have to build a data lake for it.

Or shows that you watched months ago but Netflix does not know about.

Or why you have 5 sreens of categories to choose from, but all categories have the same shows in them, anywhere around 60-75%.

I keep hearing praises of Netflix's recommendation engine. I wonder when they will roll that out! Because when I browse Netflix, I have 5 pages of categories, each with upto 75 shows. However, and this is the critical point, each of those categories contains about 60-75% the same shows, presented in different order. And with no respect to genre, to boot.