Style Guidance Official Blog

Google’s porn problem and how to fix it

Have you used Google Images lately? It’s filled with porn, even for the most innocent of keywords.

Now to be clear, this is me using it with safe search off…but that shouldn’t matter. If I search for a completely innocent key phrase, I shouldn’t be seeing porn on first page of results, regardless of whether or not I have safe search turned on or off.

I’ll give you a few examples(warning all these searches are NSFW):

• green shorts
• red top
• cute girl
• orange pants
• tank top

All of these results end up with porn on the front page of the results. Do you think a 10 year old girl searching for green shorts or a red top, should be seeing any of those images?

So how do you fix it?

Well it’s pretty easy. Just apply safe search to the first 3-4 pages of search results(except for porn searches), regardless of whether or not safe search is on. Anyone who wants to see porn for a term “green shorts” can wait till page 5 to see the nudes. And the rest of people who aren’t looking for porn when looking for green shorts, can avoid seeing porn.

But what about porn related searches? What do you do for all the people searching for porn on Google images? It’s easy, you just create a white list, add 500-1000 porn related keywords(porn, sex, soviet steel production, basically whatever term Google feels is a term that people use to search porn) to it, and then just apply it to search results.

If the search string has any of the white listed keywords, and the person has safe search turned off, you kinda know that they are looking for porn, so you show them the porn they’ve been looking for.

And if it doesn’t, you show the non-porn images that the person is actually expecting to find.

This way people won’t get ambushed by porn when searching for innocent keywords, and the people looking for porn can get their fix.

Now some will say, that if you have safe search off, you shouldn’t worry too much about porn. Yes that is true. But what does it say about Google results if they think the #2 result for “red top” should be a nude photo? When people search for “green shorts” are they looking for porn or are they more likely to be looking for photos of green shorts? When people search for “green shorts” are they looking for porn? or is it more likely that they are looking for photos of green shorts? The concern here is that Google is showing completely irrelevant results for image search, when you have safe search turned off.

Turning off safe search doesn’t mean, I want to see irrelevant results at the front of my search results.

Latest update.

The latest update has just rolled out, here are the changes that were implemented:

• Regular username/password login/registration option was added. As usual you can still use 99% of the features w/o registering, but for those who want to vote, now you can register easily.
• Every registered user can now create new tags
• Every registered/unregistered user can now edit tags
• Every registered/unregistered user can now add comments to questions/answers
• Every registered user can now vote
• You can now add/answer questions in a row. There is still a limit of 3 new questions per hour on new users, but at least now you can do it back to back, instead of waiting 20 minutes.
• Subjective question warning has been removed
• A bunch of background stuff that you won’t notice, but which should make the site perform a little bit faster.

Techcrunch just deleted all comments older than 1/24/2010

Seems like Techcrunch just wiped every single comment older than 1/24/2010.

My guess is that this has something to do with them getting hacked two days in a row.

So if you find an old Techcrunch article, dated 1/23/2010 or older…and it has 0 comments…now you know why.

Google hates Survey.io - 32% drop in SE traffic

So I’ve been reading Venture Hacks lately, and there was a post about how you can use surveys to find out what your users think. Good idea right?

So I install the Survey.io javascript snippet…and sit back and relax.

And BOOM, my Google traffic takes a huge hit. 32% drop in traffic overnight(went down from 531 to 364) just from installing a little widget.

Granted my traffic is pretty small right now, so it could have just been a spike, but it’s a pretty big coincidence that my traffic took a dip while the widget was installed, and recovered as soon as I took it off.

Update: Hiten(@hnshah - the guy behind survey.io and KISSmetrics) has modified the code(removed Google Analytics from the embed) so it should be fixed now.

Apple releases the tablet.

So today is when Apple releases the tablet.

Whether it’ll be called the iPad or iSlate or iTablet, it doesn’t really matter. Same goes double for features…as the iPhone showed us, any missing features will be provided by the thousands of developers at Apple’s beck and call.

The ONLY thing that matters is the price. Price it too high, and the revolution is over before it starts.

Basically if it’s priced at the 600-800 price point(doubt it can even be lower than that), it’ll be a runaway success, since anyone will be able to afford it. At this point, the tablet will really be a new medium that everyone is hoping it’ll be.

If it’s priced at the $1,000 price point, it’ll be a hard sell, since most people can’t spend $1K on a toy in these economic times. By going to this price point, Apple will essentially cut their potential market in half. At this point the tablet can still survive as a medium, but it’ll be a struggle.

If it’s priced at 1,500, even the most hardcore Apple fan boys won’t buy it. At that point it’ll be just a status symbol…the only people buying it, will be those who’ve got money to burn.

Another thing to consider is the price of the contract. This can boost or kill sales depending on how much that’ll be and how much that subsidizes the cost of the tablet.

Personally I think it’ll be in that first tier. The Tablet is a way for apple to redefine entire industries, they won’t want to cannibalize sales by pricing it too high.

Image is fake in case you don’t know, photoshop by gizmodo

The startup culture is filled with people with ADD

It seems like the startup culture is filled with people with ADD.

They come out with a project, see little or no traction the first few weeks-month, ..and then abandon it to go work on something else.

Everyone needs to take a moment and realize that there is no such thing as overnight success. Putting out a project, getting covered in TC, getting millions in VC, and then becoming a top 1,000 web property doesn’t happen overnight.

And even when it does, the # of companies who it happens to can be counted on one hand. Really, the only one in recent years that even comes to mind is Mint.

The reason is survivor bias. You read TC, and mostly hear about the successes. Even when you hear about a failure…it’s still about a company that was successful enough to raise money and reach some decent traction. What you don’t hear about, is the thousands of sites that didn’t even get TC coverage. What you don’t hear about, is thousands of sites that do get TC coverage, that hit deadpool.

The dirty little secret, is that starting a new website is always slow. You always start out with 0. And you always start out by getting only a couple of people a day. The thing though, is if you stick it out, eventually you’ll upgrade to 10-20 people a day. Then 100-200. Then 1000-2000. Then 10000-20000. Then 100000-200000. You just have to stick it out to survive long enough to hit those numbers.

Remember the old hockey stick, you just need to survive long enough to pass the hockey stick threshold.

I’ll give you an example. Remember twitter? This little known web app that is now the #14 site in the world according to Alexa?

Well here is their early traffic graph(source:techcrunch), this is the earliest graph I managed to find. As you can see according to the graph their December 2007 traffic was about 100K uniques.

100K uniques…that’s nothing right? Well here is another fun fact.

Twitter was founded March 2006, and launched July 15, 2006. That means it took them 6 months after launching to reach 100,000 uniques a month.

So stick it out, you never know….that little app you want to abandon after 1 month, might be the next twitter.

Obviously not everything works out that well, you need to figure out if it’s a service that people actually want, never the less, you should at least give it 6-7 months to prove itself, before you decide to kill it.

HN needs to post new stories on the front page to give them a chance

If you know Hacker News, you know that it’s a pretty big programmer/startup site.

But it seems like noone views the new tab(i.e. in the example below, not a single one of those stories got an upvote…even after 27 minutes.(this was in the middle of the day too)

What I think HN needs is to post the latest submissions on the front page(like reddit). That way every new story gets the extra exposure for a few minutes to see if anyone thinks it’s worthy. If not..it goes to die in the new tab. If yes…it has some chance to survive.

Make me shave my head

I’ve noticed that you guys really step up whenever we have a special number to hit. For example, when we were aiming for 2,000 Qs by our 2 month anniversary - you’ve stepped up and made that happen.

We now have 971 Facebook Fans, and 1582 twitter followers…and the site had 35K visits last month. So it’s time to get us another bump.

Remember, questions are the lifeblood of the site, the more questions we have, the better resource the site is. And the better the site is, the more people will come here for their answers. Which in turns helps pay for all those giveaways we are running.

So here is my challenge, post 1,000 new questions(even if they are similar to existing questions…still counts as long as it’s not a direct copy) in 1 week and I’ll shave my head. Shouldn’t be that hard, if each of you posts 1 question, we’ll hit that number no problem.

I know it’s a little WTF…but I wanted to do a challenge that was a little weird.

Oh and you are more than welcome to post about this challenge anywhere else, the more the merrier.

This week’s goal is 3,664

edit: well week 1 was a failure(only 30.2% complete). So the # resets. Hopefully the Kindle giveaway that we are doing this week will be a better incentive.

To start, click here.

-Andrew

p.s. all entries will count as usual towards the giveaways.

p.p.s. if we don’t hit 1,000 Questions this week, the counter resets, and we try again next week.

p.p.p.s. this is what I look like now:

p.p.p.p.s. Actually did it…even though you guys haven’t really hit the number

Promoting on forums is a huge waste of time

So you think you are going to market your new site by posting on forums? Hit up a few dozen big sites? A site with 100,000 users…you gotta get a few sales/users out of it. Right? WRONG!

Here is the dirty little secret. Most forums are dead. Sure they may say they have 100,000-400,000 users. But in reality 99% of those have stopped coming. You gotta realize that forums has been around for a while now, it’s pretty easy to accumulate 100K users, when you’ve been around for 12 years. It doesn’t mean any of those people are still around.

So in reality, a huge forum like that, will have ~1,000-3,000 people online. But those users will be split across all the subforums of that site. So in reality, at most, you are looking at 100-200 views of your thread before it disappears of the front page. Less if it gets deleted for spam.

And do you know how many clicks you’ll get? 10? 20? Try 5.

Now that’s if you post as a new user. But what if you are an active member of the community? Better numbers? Sure, but not by much. I’ve promoted one of my sites as a 6 year member of the community(forum had 150K users)…this kept my thread from being deleted, and it got 60+ replies, which kept the thread on the front page for a week. Result? 800 thread views, 61 clicks.

So as you can see, forum posting is a huge waste of time, you’ll spend 5 minutes getting registered, in exchange for a few clicks…not worth it!, You’ll get better results from being listed for 3 minutes on a reddit “new submissions” tab.

Do you know who forum posting works for? Actual spammers(you know the guys selling viagra). They use this bit of software called XRumer, which does all the work for them. They type in all the info, and it goes spamming all the forums in it’s list. And unlike you, they’ll be hitting 100,000 forums at a time. They’ll still be getting 1-2 clicks per site, but they make up for it with volume.

For “regular” folks, who just want to promote their new site, forums will never give you a return on the 5 minutes you spend registering.

Now some will say, oh just post a lot on the forum, and keep your site in your signature. Build up your reputation, and the clicks will follow. But have any of you actually tried it? Sure you’ll get a few more clicks this way(~50-60 a month), but you’ll also be spending a ton more time doing this. Time that is better spent elsewhere.

Disclaimer: I’m not saying that forum posting is the only promotion method you are thinking off. What I’m saying, is that you shouldn’t even consider it as one of the methods..

Site statistics for month #2.(35K Visitors, Top 100K Alexa, PR5)

So it’s now been 2 months since I launched Style Guidance.

So here is the report for month #2. I’m comparing November 18th-December 17th numbers to December 18th-January 17th numbers.

• Visitors: 11.6K —> 35K(up 202%)
• 
• Returning Visitors: 2,773 —> 7,491(up 170%)
• Daily Visitors: ~700—>1.2K (note, the ~700# was in the middle of a big promotion, the 1.2K# is just business as usual) (up 72%)
• Daily Search Engine Traffic: ~110—>~500-550 (this is my most important #) (up 400%)
• Contributors(anyone who asked/answered question): 805—> 1941(up 141%)
• Questions: 627—> 1923 (up 206%)
• Twitter Followers: 749—>1466(up 95%)
• Facebook Fans: 356—>915(up 157%)

We also had a few other milestones, which I want to start tracking for future reports.

• Page Rank: 5(not really a mile stone, since it takes like 3 months to update, but still)
• Alexa Rank Global: 101,789(but for all intents and purposes this will be sub  100K today)
• Alexa Rank U.S.: 19,677

Overall I’m very happy with our progress, to be honest, I was expecting the first 6 months to be at ~100 users a day, and we are doing 10 times that.

For next month, my hope is to hit 50-60K uniques. Since there won’t be a week long ghost town during Christmas/New Years.

P.S. As you can see I switched to a different blog template, so hopefully it’ll make things easier to read.