Posts Tagged ‘Website’

September 12, 2016


Yesterday I came across an article from Tech Crunch about google heavily pressuring website owners to move from the http protocal to the more secure https. In the article they’re discussing about chrome in the article, not search, but I think it’s a reasonable assumption that a similar policy may be in place in search, if not now than in the near future. If two websites have comparable content, why not favor the one with the better user experience?

I admit as of late I’ve been a bit more reactive and not enough proactive than I’d like. Still, it seemed like good timing to make the change since I was already going through all this renaming pain. I recently opted to remove the ‘www’ subdomain from all my URLs which appears to have affected the google index for my websites. Just as the URL is technically different from, is technically different from While it would have been better to have gone from to in one step, making the change would at mean any pain from renaming would at least overlap a little instead of costing me twice.

It would be nice to log into my blog over an https connection so I don’t have to fear a man in the middle attack any time I write a new blog post. I’m also considering eventually adding an ad-free experience for a nominal fee which would necessitate a secure connection.

The final driver towards making the https plunge was the discoery that dreamhost offers and SSL certificate for free. Yesterday I added the Let’s Encrypt! certificate to When that went off smoothly I went ahead and added one to

I have no idea if it’ll help my search rankings, but at least it couldn’t hurt. Probably.

February 5, 2016

A New Look

It finally happened. I got bored with my blog layout. I have never really liked the narrow format, but stuck with it because I enjoyed the header photo. Nearly two years have passed since I had taken that photo and it was filling me with more guilt than joy. I felt guilty that I was using one child’s photo and not the other, even though the time stamp of the image was the only way anyone would ever know which child the photo was of. It felt beyond time for a change.

The new layout isn’t much different from the old. I traded in the blue theme for a mostly black and white one, and adapted my blog tag line to reflect changes in lifestyle. For now I’m header photo-less, but I intend to change that with a self portrait as soon as I’m able to find the time to play with my camera and get a halfway decent photo I don’t hate.

I’ve been experimenting more and more with my camera phone. The end goal for my business is to move beyond webapps and into photography apps for the phone. I’m sure that when that happens I’ll want to share photos more publicly than I’ve been doing. At the same time it didn’t really feel right to open up my personal Instagram account, with all those kid photos, to the public. I post kid photos publicly here, sure, but there far more selective and curated. And small. I never upload anything beyond a 400px dimension photo. No accidentally readable street signs in my photos! I decided to create a second instagram account.

Since I don’t have a business identity yet, the new account is sarahktyler. I pledge that all photos uploaded on that account will be taken from my phone. Hopefully some time in the not too terribly distant future they’ll have been taken from my app, rather than the native camera app.

July 17, 2015

Name Uniqueness Analyzer

It lives!

For a while Uniqueness of Baby Names was one of the top blog posts. It even got pinned on pinterest. But as fun as it was writing the post (and doing the math!), the information wasn’t interactive. I hoped it was a fun read, but that was it. I wanted the math to live on.

I’m pleased to announce I’ve turned the post into an interactive web app. Simply enter a name to see how popular it is for a given year. The name uniqueness analyzer will also tell you the odds of encountering another person with the same name. If you’re searching for a name, the name analyzer can also suggest names based on how unique you’d like it to be.

This is my first webapp launched since leaving Google and deciding to start my own start-up. While the main start-up idea is still baking, I thought I’d launch a few apps to both keep me coding. I’m both pleased and embarrassed by how long it took me write it. On the one hand, I left Google a little over two and a half months ago. That’s a really long time to launch anything! On the other, I have Alexis home during the days two days a week, and lost a full day dealing with the death of the washer and drier. From “I will do this” to “It launched!” was only two days.

I’ll be watching the Name Uniqueness Analyzer closely to see what kind of adoption it gets. The dream is to launch enough of these web apps to replace my grad school salary, freeing me up to work on my start-up without worrying (too much) about the finances.

Egads, I missed my blogversary again. Time is just racing by these days.

My metrics this year are basically flat, or down slightly. Not too terribly surprising since I’ve been either working or on maternity leave with less time to blog these days. I didn’t even write as much about my Ziggy Pregnancy as I did with Zippy. I’ve also been mostly sticking to my promise to blog less about Nicole, although I clearly miss it.

My most popular post by such a big a factor of 18x is my statistical model for labor. I really must follow it up with some more geeky posts. I’m kind of surprised the baby name uniqueness post isn’t more popular, but there’s also not a fun tool accompanying it. I am thinking about changing that! Stay tuned.

July 4, 2014

Technical Difficulties

I’ve had a string of technical difficulties lately, some with regard to my website, some with regard to my on-going backup efforts.

On Monday my webhost upgraded the version of PHP running on the server. Unfortunately one of the plugins (not the one I wrote, phew) I was using was incompatible with new PHP. It was relying on a technique that was so insecure, the new compiler catches identifies the problem and refuses to execute the script. The problem is, this effectively brought down my blog.

Even though I had plenty of warning, I didn’t get around to checking my blog until yesterday.

June/July Traffic
Ouch. Those stats do not make for a happy Sarah.

At least it’s back up now. The offending plugin has been removed.

The other technical battle I’ve been fighting is getting my backup system setup.

The first problem was with our ISP. Our download rate dropped to 2.5 mbs (megabytes per second), a mere 5% of the promised rate. As much as I would love to blame my ISP, I’m not sure it’s entirely their fault. Living in an apartment in the heart of silicon valley, I’m sure there are a number of start-ups within 50 feet of our apartment. One of them could be saturating the line. Of course the onus is still on the ISP to deliver what’s promised, so they’re not off the hook either.

What does a cruddy internet connect have to do with transferring files on a home network? I’m glad you asked that question! Apparently in order for our home network to work (also set up by our ISP), our modem needs to connect to the outside world. That’s a bit like living on an island, and driving over the bridge to the mainland and then coming back any time you want to go anywhere else on the island. It’s that stupid. With our connection being so slow it would time out, and with the bridge to the mainland down, those files weren’t going anywhere.

The second issue I had was with crash plan itself. I’m using multiple external hard drives for my backup. All hard drives have a UUID (Universally Unique Identifier). It’s a way of identifying the drive. Crash plan identifies a drive not by it’s UUID, but by the drive letter the disk has been mounted to. The problem is the drive letter can change and is usually assigned in order of the drive’s being mounted. A system restart can change those drive letters! That’s like running a pizza delivery place and identifying your customers with phone in orders based on their arrival time rather than their name.

This last one was fixable by manually assigning the drive letters, it just took a while to realize what was wrong. At least I’m making progress. I finished sorting through one external hard drive. Four more to go through.

In my never ending quest to learn about the business of blogging I recently stumbled onto FTCs rules for .com disclosure. I had previously setup a disclosure page, and often mentioned the types of revenue from my blog (namely Adsense and Amazon Affiliate links), but I realized I could do better.

The basic premise from the FTC’s release is that any visitor should know a link is sponsored, before viewing said link, regardless of what device they’re on.

With roughly 300 posts I didn’t relish the idea of labeling every post or link by hand. Manual processes like these are prone with errors, and I was sure to make one. A better solution was a plug-in to automate the process. Since I only use one affiliate program, it’s a pretty simple to automatically parse my blog posts looking for affiliate links. Alas, the only affiliate plug-ins I found were designed to cloak (i.e. hide) the fact that a link was an affiliate link – the exact opposite of what I wanted to do! I finally had an excuse to write my own plug-in.

Writing the plug-in ended up being easier than I thought it would. The harder question was coming up with a labeling scheme I liked. I didn’t want anything to interrupt the prose of the post. I decided to go with an “affiliate link” title to all affiliate links’ <a> tags. Now when you hover your an affiliate link, you see a little pop box that properly identifies the link. I’ve seen this method used before, namely by I’ve verified the title also appears on my blog posts in, and

There is just one problem.

The devices used to view my blog in 2013

Over half of my web traffic is coming from a mobile device or a tablet and there is no concept of “hover” on such devices. Affiliate links are still unlabeled on these devices. Since I wanted to minimize the time I was not compliant with the FTC, I decided to bite the bullet and add a disclaimer to the top of each blog post using my plug-in. I hope to come up with another solution in the future.

Side note: Apparently there’s a belief out there that search engines will demote pages with affiliate links, hence the prevalence of affiliate link cloakers. While I have no first hand knowledge of the truth of this, it feels like an old wives tale to me. In fact, cloaking is probably counter productive. It’s trivial for a computer program to find where a cloaked link leads. If there’s one thing I’ve read about SEO time and time again, it’s engaging in deceptive practices, like cloaking, will eventually negatively impact your ranking.

February 9, 2014

Mobile Inclined

In what I am sure is mostly boring website news, the first cut of a mobile theme for my blog is now live.

One of the nice things about wordpress is the availability of plug-ins. There’s a plugin for just about anything. Problem was, non of the plugin’s did exactly what I wanted. I decided to go with mobile-smart since it got me the closest to where I wanted to be.

The key functionality I wanted included:
– Have two themes, one desktop one mobile ( ✔ )
– Allow mobile users to switch to the desktop theme ( ✔ )
– Allow mobile users to switch back to the mobile theme ( ermm…)

So what’s a computer scientist to do? Crack open the code of course!

Original code (paraphrased) for creating links to switch between themes

$is_mobile = $this->switcher_isMobile();
if ($is_mobile || $options[‘allow_desktop_switcher’]) // (1)
  if ($is_mobile) { // (2)
    … //(code to write button/link to switch from mobile to desktop)
  } else { // (3)
    … //(code to write button/link to switch from desktop to mobile)

The function $this->switcher_isMobile(); returns true if the user is currently viewing the mobile template. The parameter $options[‘allow_desktop_switcher’] is a flag indicating whether any user is allowed to switch to the desktop version from the mobile version.

Since I don’t want desktop users switching to the mobile theme, I need to set that flag to false. That means the only way to enter the first if clause, (1), is if $is_mobile is true. But then there is no way to enter the else clause, (3). A link to switch back to the mobile version is never created. A mobile visitor can return to the mobile version, but only after shutting down their browser and starting a completely new session.

My Change

$is_mobile = $this->DetectIsMobile();
$is_mobile_shown = $this->switcher_isMobile();
if ($is_mobile || $options[‘allow_desktop_switcher’]) // (1)
  if ($is_mobile_shown) { // (2)
    … //(code to write button/link to switch from mobile to desktop)
  } else { // (3)
    … //(code to write button/link to switch from desktop to mobile)

Green indicates changed lines

In the above code, $this->DetectIsMobile(); returns true if the visitor is using a mobile device, regardless of which version the user is currently viewing. Now, the if clause at (2) is entered if the visitor is viewing the mobile version, and the else clause, (3), if viewing the desktop one.

It’s a very small change, almost not worth a blog post. I know at some point I’ll update my plugins and probably lose my edit. At least now I can easily find & redo what I did!

November 10, 2013



Last week I started my new job at Google. I’ve been kind of mum about the specifics here (and it felt really awkward to not answer the question when asked directly, my apologies for the clumsily dodge.) I wanted to hold off on sharing the news because I wasn’t sure what changes I would need to make to my blog. I didn’t want to guess incorrectly and land myself in hot water before even starting.

I have always viewed my blog as a chance to grow my ‘brand’, a sort of social complement to my online resume. I hope to grow my reputation through my blog and as such I often write about the internet which has on occasion included my views on search. My dissertation and past publications are all on personalized search. Clearly this research was done outside the scope my Google employment and should be fair game to write about. On the other hand, Google is also interested in Search and if I post something related to my prior research it’s possible someone might misinterpreted the post as relating to Google’s research. To be clear I have never, and will never, discussed the details of my work in my blog, it’s the appearance of doing so that I want to avoid. For now, I will have to content myself with the ability to put ‘google’ on my resume for search expertise cred, and avoid blogging on the topic.

As an aside I hope my blog isn’t branded as a ‘googler’s blog’, just as I hoped it wasn’t branded a ‘mommy’s blog’. I am a Googler, I am a mom, but I’m also a budding photographer, math nerd, science junkie, bargain hunter, etc.

Another issue is that I use Google Ad Sense. I’ve written about it in the past, and had some fun playing with the numbers. I personally find the topic of blogging revenue fascinating, and I had hopped to continue writing on the topic since there are so few detailed articles on the subject out there. Alas, this area is now obviously off limits for me.

It’s a worthwhile trade for full time employment doing what I love. I can tell I’m really going to enjoy this next stage of life, if I can ever get around to unpacking those darn boxes!

September 2, 2013

Back to Blogging Basics

my baby you'll be
“I’ll love you forever
I’ll like you for always
As long as I’m living
My baby you’ll be.”
– Robert Munsch

When I was pregnant with Nicki, and she was a tiny embryo without a nervous system or brain capable of cognitive thought, I felt like I was sharing my story. Even when she a newborn in the “fourth trimester”, crying was her only means of interacting with the world and was completely dependent on Domingo and I. Her story was still our story. Then she learned to hold and manipulate objects. She learned to crawl, and walk and run. She’s talking. She has opinions, and gets frustrated when we don’t understand. Her personality is shining through in spades. While our stories are still intertwined, I’m becoming more and more weary of accidentally stepping my bounds and sharing her story.

I blog under my real name, without a pseudonym to offer at least a layer of indirection. By extension Domingo and Nicki are being blogged about in their real names as well. While I can ask Domingo what level of sharing he’s comfortable with, Nicki will not understand the possible ramifications for quite some time yet. I have so far used myself as a guideline (“Would I be comfortable if my parents posted this story about me?”), but I tend to be more open than the average person. I’m certainty more open than Domingo, and Nicki may take after her daddy.

I’m also concerned that what I might blog about would negatively impact her later in life. Stories of baby antics are a potential source of embarrassment but are unlikely to do real harm, especially since they’re pretty universal. Yet what if I accidentally hint at a health issue or a learning delay? Could that impact a potential future employers hiring decision, even subconsciously? I am dyslexic. It’s something I’m comfortable about enough to share openly, but there have been people (a high school teacher and an ex-supervise come to mind) who mocked me for it. If Nicki should find herself in a similar situation, it should be her choice to share it, not mine. Delays and health issues are pretty obvious topics to steer away from, but blogging is still relatively new, we don’t necessarily know the ramifications discussing seemingly innocent topics may have on our children.

A few years ago Kanses state published a study on societies impressions of working moms. In the study researchers showed a video of a mother and child interacting to two groups of people. In the first they described the mother as a stay at home mother. In the second she was a working mom. There were no other differences in the descriptions and the same video was shown to both groups. Not only did participants judge the “working mother” more harshly, but they had a more negative view of the child. On the one hand this study’s findings aren’t terribly surprising. It’s a form of confirmation bias called biased interpretation. We interpret our surroundings to fit our beliefs. If we believe the working outside the home damages the mother-child relationship, we are more likely to view those around us who work as having impacted relationships with their children because it fits our world view. Still, this study has been haunting me lately, particularly because the results extend to the child. I’m a working mom. I’ve formula feed (I’ve breast fed beyond a year as well, but that’s usually viewed as a positive by society). I’ve employed cry-it-out. I’ve even admitted to breaking the no media before two rule. I’ve admitted all these things here and while I absolutely think these were the right things to do for Nicki, I know they are somewhat controversial. Could you, anonymous reader, be subconsciously viewing Nicki more harshly because I blogged about those things openly?

So I’ve been making an effort to share less about Nicki since she turned one. I’m posting less candids, I’m sharing less antics. There is no way to completely disentangle Nicki’s story from my story, Nicki will be less of the focus. I’ll be returning to my original intention of blogging – a place to practice writing and maybe even growing my professional brand.

While I think this is the right decision, it wasn’t an easy one. I have a fear that Nicki will one day grow up, read my blog and wonder why I talk about her less. Will she think I loved her less as a toddler than an infant? Domingo and I will likely have at least one more child. Will the next child (and by induction nth child </math joke>) think I love them less if I don’t share as much about their first years? And if I share their first year as much as I shared Nicki’s, will Nicki feel less loved that I am discussing her less at the same time?

Domingo and I have a running joke that he knew I loved him when I was willing to accept the marriage tax penalty. Hopefully, one day Nicki and I will have a similar joke, that she knew I loved her because I was willing to blog about her less.

February 28, 2013

2nd Year Blogiversary

I’m not sure how I missed it, but my 2nd year Blogiversary was 4 days ago! You’d think with all my posts on meta-blogging I would have noticed it coming. I blame my distracted state on prepping for upcoming interviews.

So how’s my fledgling little blog doing?

* In terms of traffic sources, I had five times as many visitors from Pinterest last year than the year before!
* I have ten new pins, bringing my total to fourteen pins!
* Overall, page views is up 139% from the same time last year!

I noticed the newborn photography was the third most profitable page. I admit when I saw that I envisioned a random surfer thinking “I’ll try this do-it-yourself stuff”, stumbled onto my blog, and think “no way! I’m hiring a professional.”

The most popular page is my Labor Predictor. On any given day about 40-50% of the page views to my blog are on the labor predictor. I’m glad it’s so popular, it was fun writing it! I love my math-y posts.

My goal for the coming year: increase the number of non-mommy related posts. I was talking to someone the other day and it was clear that he thought of my blog as a “mommy blog”. Obviously being a new mom is a major part of my identity right now, but it’s not the only piece. While I’ll never shed the mom title, Nicki will grow older and more independent. I plan on keeping my blog for a while, not just while I have young kids!

With that said, I’m off to go take more photos of Nicki. Because I’m an obsessive momtographer like that.

Older Posts »