ironkeith

There are no comment on the blog, so I'm just going to put this here and hope the author sees it (as they appear to have submitted the post):

You don't want to write SQL in your controllers. It's a recipe for future pain and suffering.

- If a second action needs the same data, it's likely that you'll end up duplicating your query. Especially if you work with other people, and they don't know about the queries that exist in all of the different actions.

- If you change the structure of the data in the DB, you need to find every affected query, in every action. It's extremely fragile, and prone to breaking.

If you have a model, with a nice public API, that interacts with the DB, you know that you've isolated the change to just that one place. All of the controllers that call $model->some_data(); will continue to work no matter how your change your data source so long as you obey the API.

There are million different ways you can approach that, but I strongly recommend that you find one that works for you, and stick with it.

I genuinely like the software, and they emailed me a $10 off promo the day before my trial expired. The entire process leading up to the purchase was top notch.

Oh, and I didn't know there was a free, quality alternative until shortly after. A painfully bad checkout is the price I pay for being too lazy to do any research. ;)

I recently bought Parallels, and it wasn't much better. From the email I sent them (after I finally found an email address):

- When I arrived on the checkout page (from the link in the nice email you sent), you had tacked on a $8 digital backup fee. That is a complete bullshit charge, and you added it by default with no explanation as to what it was, or why I would want it. It was also not particularly obvious whether or not I could remove said add-on. It is a shameless, low class, money grab.

- I switched my pricing from USD to CAD, and you added $10 despite the two currencies trading at parity (I quickly swapped back, and just paid in USD). Another shameless, senseless money grab.

- You require my name, home address, email address, and phone number in order for me to purchase downloadable software via Paypal. You do not need any of that information. You want it so that some tool in marketing can have pretty powerpoints. It should be optional. (Making it required simply requires me to make up information; a waste of both our time.)

- The “send me email spam” checkbox was checked by default. The only reason for this is to get permission to spam people too lazy to pay attention to the checkout. It’s another tasteless scam formulated by a greedy, customer hostile executive tool.

- After purchasing, I entered my serial number, and was informed that I would have to register to receive updates to the software I just purchased. I cannot explain to you how incredibly asinine I find that policy. 

- The registration process once again required information I didn’t want to give you (requiring me to give you fake information)

- After registering, and confirming my registration via email, and entering my email/password combo into the application I was told that I had given invalid credentials, and to try again. (I hadn’t. I’m pretty good with copy/paste.) So I guess NO UPDATES FOR ME?

One small thing, on this page: https://www.appointmentreminder.org/pricing

The strings from the "hot" tag to the "Small business" column are wrong. One of the strings needs to come from behing the column, otherwise it looks like there's some weird bug sitting on the column, holding a "hot" tag.

See: http://imgur.com/gjVbb

I tried that link on my phone and got redirected to the "mobile" home page, then couldn't find the article in question. I'm pretty sure people from ars are in here, if so: please fix this. I read 90% of my news on my phone during down time. Thanks.

I laughed at the reply to "cat".

In the book "Superfreakenomics", IV is shown to be actively developing insanely simple/improbable solutions for hurricanes (by cycling warmer ocean water with cooler water using tires and concrete), global warming (by pumping sulphur into the statosphere), and malaria (with female mosquito detecting lasers). The book protrayed them as developing prototypes for those "inventions", so patents seem fitting.

One last, really minor thing:

When I create a new event, and click "Untitled Event" to edit it, it would be really nice if it auto-selected the text for me. As is I had to click, then double-click, so it's three click for a one click operation.

Maybe this is just me, but I couldn't get the site to work. Check my profile for the email address I used to sign up.

First try, Firefox 3.6 on Snow Leopard: - Opened iPhoto and dragged a photo in, progress bar filled, but no picture appear, and no indication of success was given. Check file info and realized it was a NEF. - Repeated process with a JPG. Progress bar filled, but same results.

Next try, Safari 4.0.4: - Clicked upload button - Selected two files - Progress bar filled, but no pictures appeared, and no indication of success was given.

Now when I visit the home page, it says I have two events, but neither of them are clickable because they have no pictures.

Beautiful site though, I was really impressed by the GUI (if only it worked...).

Another example would be Wolfram|Alpha. The site is free, but the iPhone app, and API access are quite expensive.

Apple recently released the Gallery app, and released a new version with an authentication bug fix the next day. It's pretty safe to say that they don't have to resubmit apps for approval.

That's true, I supposed I'd never considered what it would be like from a publisher's perspective. Does HN push enough traffic to make it worthwhile to create content directly targeted at its users? Knowing what to expect for traffic would help me evaluate if its worth it to invest 2-3 hours at a x% chance of hitting the homepage.

So far as the value of 'x%' goes, there are certainly ways you could game the current system in your favor: linkbait the title to appeal to a very specific demo, get a few friends lined up for a quick upvote... it doesn't seem like it would be too much trouble. I often see very recently submitted articles with 3-4 upvotes at or near the top of the homepage. If the only reason you're writing is for traffic, there certainly appears to be a lot of opportunity. Am I wrong?

I've submitted 18 articles, and 9 have made it to the homepage (http://news.ycombinator.com/submitted?id=ironkeith). I am certainly not a power user, and I don't have a circle of friends who quickly upvote my submissions to game the system. So far as I can tell, people use the new page, and interesting articles make the home page.

That said, I also think it's really important to ensure that the title you submit is properly phrased. The articles that have received the most votes have inevitably been the ones I put some time into rewording to appeal to HN users.

You should be recycling, you're wasting valuable nickels.

I guess Dropbox ended up taking legal action against dropbox.com when they started placing ads for Dropbox's competitors. As a result, the domain holder ended up handing the domain over (no mention of purchase in the article).

http://www.techcrunch.com/2009/10/13/dropbox-acquires-the-do...

Yes, but prior to that they used getdropbox.com, not dropbox.net (or something of that ilk). I do believe they were able to lawyer a squatter into releasing the dropbox.com domain after they were proven successful.

I'd buy that.

I got super paranoid about that, so I created a system for creating quasi-random passwords for different sites, but based on the same rule:

salt+[reversed first four letters of site name]+[number of digits in site name] not the actual rule I used, but you get the idea

That was okay, but kind of annoying. Plus, I figured that the rule wouldn't be overly difficult to break, and then I'd be just as screwed as if I used the same PW for all sites.

Now I use 1Password to generate and store all of my passwords. I use dropbox to sync it across all my computers, and if I log in to dropbox, I can access a web interface. There's also an iPhone app, so it's not completely annoying never knowing any of my passwords, and I don't have to worry about one site storing my PW plain text and being exploited. [Now I just need to worry about my dropbox account getting hacked... here's hoping they don't store in plain text ;)]

I don't have anything to do with 1Password, and there are a lot of other apps out there that do the same thing.

While we're being sticklers for the guidelines...

If you flag something, please don't also comment that you did.

I don't know about this; it doesn't appear to support authentication, or allow me to upvote/downvote (I get that it probably can't). My only gripe with using HN in Mobile Safari is that sometimes when I go to upvote/downvote I press the wrong one (my thumb lacks precision). This doesn't solve that problem for me, so I can't really see myself spending $2 for increased padding.

I guess he's never heard of monster.com or 37 Signals.

Funny, prior to the parent post (which was really bad) being deleted, I had plus 2 karma here. Now I have minus two. I wonder if people don't get that I'm replying to a deleted comment and not the submission.

Is that a meme I don't know, or did you accidentally post to the way wrong site?

Thanks Papa, and to your brother who didn't make it home.

I have been a little obsessed with this idea lately, so I'm pretty excited to see some progress. "Training" cars to share the efforts of acceleration/deceleration, cut wind resistance, ideally reduce collisions, minimize the effects of "traffic fluid dynamics", and allow me to play on my laptop during my commute is technology I can get behind.

Sounds like a great opportunity for someone to SEO optimize a site, paraphrase WSJ articles, put some ads in and let Google index them (like already happens to dozens of blogs). Then instead of blaming Google for stealing their money, News Corp will have to blame the dozens (hundreds?) of spammy reposters who appear to fill the void.

My guess would be a CPM somewhere around $0.15. It would appear that he was correct regarding traffic though: http://siteanalytics.compete.com/wsj.com/

I'm curious as to how they picked the list of cities. By population?

"The dot coms I have known were all doomed. Almost all of them were the classic two-founder startups going after a niche in the market. All of them needed outside money to achieve their goals, and in every case the outside money wrecked the company. Uncontrolled hiring wiped out the competence and the culture of these places within months. Executives who came in with the outside money were out of their depth and resorted to arbitrary decision-making and tyranny, and sometimes deliberately failed at their fiduciary responsibilities. "

I worked for a company matching that description precisely, and can attest to the fact that it really sucked, a lot. The strange thing for me was that it was a subtle shift from "wow, this is so exciting" to "I can't believe how much this sucks". I can't even pinpoint exactly when it happened.

In the beginning it was exciting. Lot's of money floating around, everyone assuming things would be wildly successful and we'd all be rich. Reality was a hard pill to swallow (I was the fifth employee of a company that grew to just over 100 then contracted to around 25). That said, I'm not really tempted to go to the corporate world. I can't just unplug my ambition and accept being a small cog in a giant machine. I would much rather take all the lessons I've learned on "what not to do" and channel that into a new (and hopefully better) startup.

All you need to do is wait a little while for ICANN to support custom TLDs then convince PG to spend a hundred grand and all your dreams could come true (or at least the one).