Jason Lotito - Latest Comments

Re: How to get an invite to Forrst

_thp — Thu, 03 Nov 2011 10:46:17 -0000

http://forrst.com/i/B7u
Here's my application, I would love to be part of this community.

Re: Comments

Jason Lotito — Fri, 28 Oct 2011 13:29:38 -0000

This is of course just the first step. You could add methods specifically for each of the checks. This article, however, was merely focusing on using comments as an indicator. Other indicators I will talk about are conditionals and loops. The contents of an if or loop block is an indication of moving something into it's own method.

The idea is simple: when you do this, it should raise a flag.

As you mention, we should question every line of code we write. This hopefully helps to provide solutions to these questions.

Re: Comments

Eric Hogue — Fri, 28 Oct 2011 13:19:42 -0000

I like the idea of questioning your code every time you feel the need to add a comment. I think we should question every line of code we write, but comments are often a great indicator of problems.

What do you think about going even further in the refactoring? I think I would add two more methods. Something like matchEmailFormat() and matchUsernameFormat(). It would make it easier to reuse, modify and read. If someone really needs to know how you check for it, he can go in the one line methods.

Re: How to get an invite to Forrst

Starseed5 — Tue, 16 Aug 2011 14:54:42 -0000

Hi, thanks for the article. I would appreciate some feeback on my application if anyone has a chance: http://forrst.com/i/CNz I'm quite confident I left some constructive feeback.

Thanks x

Re: How to get an invite to Forrst

Squrk — Mon, 25 Jul 2011 09:19:23 -0000

Hi,
I applied for a Forrst invite would appricate any members having a look at my application.

Link: http://forrst.com/i/CBj

Thanks for reading.

Re: privacy.txt

سيد مهدي مصطفوي — Sat, 16 Jul 2011 05:20:13 -0000

Thanks so much for sharing

Re: Eclipse Color Themes

doxigo — Fri, 15 Jul 2011 12:00:17 -0000

Yes, Sepehr Lajevardi is such awesome developer

Re: What I know about designing credit card forms

Martin Wharton — Tue, 12 Jul 2011 12:55:50 -0000

Great article Jason. Full of golden nuggets. Cheers pal!

Re: How to get an invite to Forrst

DirTek — Thu, 07 Jul 2011 14:58:17 -0000

Hi there!
I also applied for a Forrst invite a while ago but failed in getting any votes. I think that my feedback was quite detailed and to the point, but I may be wrong.

My application link is: http://forrst.com/i/5l7

Cheers!

Re: How to get an invite to Forrst

mr_tim — Mon, 27 Jun 2011 07:32:10 -0000

I'm also trying to get some votes for my application - http://forrst.com/i/5oR

Re: How to get an invite to Forrst

Sl1dr — Mon, 13 Jun 2011 07:52:59 -0000

http://forrst.com/i/52l oops i put a full stop in the first link.

Re: How to get an invite to Forrst

Sl1dr — Mon, 13 Jun 2011 07:51:43 -0000

Here is the link: http://forrst.com/i/52l. I am only starting out with HTML and CSS, but I really enjoy it and am very passionate about it. Forrst seems like such a great learning community!

Re: How to get an invite to Forrst

Jason Lotito — Sun, 12 Jun 2011 22:50:03 -0000

What's the URL to your application?

Re: How to get an invite to Forrst

Sl1dr — Sun, 12 Jun 2011 21:38:04 -0000

I would love to be a part of the community but unfortunately my application is not getting any votes!

Re: How to get an invite to Forrst

Jason Lotito — Thu, 09 Jun 2011 11:25:37 -0000

Shoot me an email at jasonlotito@gmail.com and I'll hook you up.

Re: How to get an invite to Forrst

superdeluxesam — Thu, 09 Jun 2011 11:06:05 -0000

Hi Jason,

Thanks for the post; I'm interested in joining Forrst myself so it was nice to here a bit of advice about how to do the application right. If you had the time, I would appreciate some honest feedback about my current application - you can see it here http://forrst.com/i/5rx

Re: Crazy predictions for the next 5 years

Anonymous — Thu, 05 May 2011 13:27:38 -0000

Interesting that you don't consider Linux as an OS, Ubuntu, CentOs, etc are distributions of Linux but act as an OS

Re: Sometimes I amaze myself

Jason Lotito — Mon, 02 May 2011 20:55:56 -0000

You know, this is awesome. I love the internet. Things like this just make it all better. I really hope it works out for the guy.

Re: Sometimes I amaze myself

Danielle Morrill — Mon, 02 May 2011 19:40:04 -0000

Thanks for posting this, a Google Alert pointed me to your post and I might not have found the Hacker News thread otherwise. I think your advice to your friend is solid, and I am going to try to get in touch with him about a job (I work at Twilio).

Re: Normalisation of data

Jason Lotito — Mon, 04 Apr 2011 15:07:05 -0000

Very true. And I hope no one assumes that it's simple to implement. Keep in mind I did suggest that you verify with the user the correctness of the email address. If someone types in 'jasonlotito@gmail' for example, I wouldn't assume 'gmail.com' and put it into the database straight away. Rather, I'd fix it, and validate it with the user. "We noticed your email address was incomplete. Is this correct?" That sort of thing.

However, beyond those extremes, a lot of normalising can be done to help improve the user experience. We shouldn't be afraid to work on that.

Re: Normalisation of data

Jason Lotito — Mon, 04 Apr 2011 15:03:32 -0000

I've changed the title at your suggestion to clarify the topic. I've also added a bit to the beginning making it clear normalising data is not something you just *do*. It's fairly difficult, and usually less straightforward then validating and filtering data.

On a side note, I'd been working with Filters in Zend Framework lately that normalise data, which is where the idea of the name came from (besides a play on Brandon's title).

Thanks for the feedback in more than 140 characters. =)

Re: Normalisation of data

Chris Shiflett — Mon, 04 Apr 2011 14:36:49 -0000

There's nothing wrong with advocating for format-independent validation or case insensitivity. If (123) 456-7890 is a valid phone number, then so is 123-456-7890. No one's going to disagree with that.

I do wish you would change the title of this post. I expected an explanation of the subtle difference between validation and filtering, and if I didn't already know the difference very well, I would be left utterly confused.

Maybe make the title something about why stricter is not always better. Your examples are good ones, because they demonstrate a strictness that applies solely to the format of the data. That's not only unhelpful, it's annoying to users for exactly the reasons you mention.

Re: Normalisation of data

Jason Lotito — Mon, 04 Apr 2011 13:57:20 -0000

Normalisation is a better word, but I disagree with it being frowned on. Going back to the Credit Card example, penalising a user because he didn't change the default option of Visa and entered a MasterCard number is frustrating. Normalising a phone number is better than demanding a user enter the correct phone as it would be dialled from a call center in India, for example. Normalising a username or email address to a lowercase string is better than denying a user entry because their phone auto-capitalises the first letter.

Taking data and filtering out the parts that don't make sense and having it make sense is what computers are for. Indeed, I don't suggest doing this for every occasion, and not something you should apply blindly.

The examples I posed (phone numbers, credit cards, emails, etc) are all real world problems I've had to deal with. Pushing the responsibility back onto the user might be the easiest way of doing things, but doesn't encourage better software.

Re: Normalisation of data

Vid Luther — Mon, 04 Apr 2011 13:53:48 -0000

"Their are lots of assumptions you can make that will make things easier for the user. Go ahead. It’s okay." I'm not sure if that's such a good idea. Unless you know your userbase very very very well (3 verys)..

Assumptions the app makes are "magic" when they work, and the app is broken when it's not, normalizing google.co to google.com could be a problem, as the email address may be google.co.uk, or something else. This level of assumption should be done on a case by case basis, once you know your audience very well.

Just my inflated $0.02 (USD :))

Re: Normalisation of data

Chris Shiflett — Mon, 04 Apr 2011 12:08:03 -0000

"Filtering is the art of taking bad data and turning it into good data."

You're confusing filtering with normalization, which is generally frowned upon. It's best to make sure data is valid (validate) and only allow valid data (filter) than to try to modify bad data to make it good (normalize). Normalization flaws tend to create worse vulnerabilities than filtering flaws.

There's a really good book that covers this specific topic, but I'm blanking on the name of it.