If some part of your body itches or burns, you’ve probably searched for free medical advice on Google. But an overload of results can have you mistaking an allergic reaction for an STD — or vice versa. Either way, it’s no fun.

Over the next few days Google is rolling out a feature called symptom search, which is designed to show better results on the Google app for iOS or Android. A search that includes a description like “child with knee pain” will return a list of related conditions.

Typing in simple symptoms like “headache” will show a general description of your problem, options for self-treatment, and suggestions on whether or not you should go to see a doctor. Many search results will show you a condition card, which Google launched last February, and these may or may not include illustrations. Other search results will appear as cards that you can swipe on or drop-down menus that you can tap to see more information.

Google says that about 1 percent of its searches are symptom-related, so it wanted to do a better job of sending people useful results. The company created this list of symptoms by turning to its web results to find health conditions, then compared these with the medical information from doctors that it uses for its Knowledge Graph. Google also got help from experts at Harvard Medical School and Mayo Clinic.

In the future, Google plans on expanding this symptom search from US- and English-only to other countries and languages. The company also hopes to bring this to the desktop browser in the future. course, numerous companies already offer symptom-checking apps, so Google may have a tough time getting some people to switch from long-time habits with WebMD or Symptomate.

Self-driven cars use computer vision, but what about your baby monitor? Today, a company called Nanit launched with the promise of using computer vision and machine learning algorithms to watch, study, and learn from your sleeping baby. This concept is in contrast to products that make parents put sensors on their child — like Owlet, which uses baby booties, or MonBaby, which uses a button that attaches to baby clothing.

The Nanit camera captures video footage of your child from a tall, white stand that hides its power cord and leans against the wall. Its fall-resistant design is made to thwart grabby babies, and it has a built-in nightlight on its ceiling-facing top side. (If the stand doesn’t work for your nursery, you can also use Nanit on a tabletop.) But Nanit’s most differentiating features work on the backend as its algorithms analyze everything the camera sees.

The Nanit app shares insights based on your baby’s moves, like when sleep actually begins and when parents come into the room, as well as a crib heatmap of where your restless kid moves throughout the night. It congratulates parents on milestones, like the first time a baby puts herself back to sleep without a parent’s intervention. Each morning, the Nanit app shows a highlight reel of the previous night, along with a sleep score. And it gives behavioral analysis and parenting tips based on what the camera observes.

“It’s not simple to make parenting tips universal because there are a lot of parenting approaches out there,” says Dr. Assaf Glazer, Nanit’s CEO and co-founder. “Just giving them awareness is 50 percent of the way.” He spent the last 15 years studying computer vision, earning his PhD at Technion in Israel and his post-doctorate at Cornell Tech in the field of machine learning and computer vision. About five years ago when Glazer had his first child, he was curious about his son’s sleeping and behavioral patterns but found existing video monitors frustratingly bad. So he did what a lot of tech CEOs do: he made his own with help from co-founders who had worked at Philips and Applied Materials.

Like other computer vision and machine learning devices, Nanit gets smarter and more personalized over time, and Glazer has high hopes for what this camera and its analytical smarts could do in the future. “Our goal is to expand the boundaries of human observation, and we start with babies,” he says.

Nanit will cost $349 when it ships in September, but it’s available for preorder starting today for $279. Along with the camera, you’ll need to pay for a subscription to Nanit’s analytical software. People who preorder get the first month free (each additional month costs $10) and can save 50 percent when they buy a full year of the service all at once.

If you’ve been dreaming about owning a customizable smartwatch, the modular watch that London-based Blocks promised is now available — but just for pre-order. And nothing gives a pre-order announcement a little extra spice like a teaser video:

This watch, we we told you all about last fall, was originally expected to ship in May, but pre-orders will now ship in October barring any more delays.

The Blocks watch costs $330, and comes in onyx black, sunrise red, or marble white. That price includes four modules that you can swap in and out on your wristband, and you can choose which module you want from six available options: Extra Battery, Heart Rate, GPS, Adventure (a sensor that measures temperature, pressure, altitude, and humidity), Flashlight, and Programmable Button. Additional modules cost $35 each.

The idea of modular anything is a hot topic right now, especially Google’s Project Ara, which is also expected to ship with six slots for modules when it comes to consumers next year.

I should’ve known it was too good to last. It started out slow, as the best relationships often do, but I quickly found myself needing him more and more. Whenever I reached out, he was there without questions or judgement — on late nights, when no one else showed up, and even when I forgot my wallet at home. It just worked. But a few months ago, he wanted other people to join us, and I refused, which made everything more complicated. Now I feel nervous and confused about contacting him.

I just want things to go back to the way they used to be, when my Uber app simply sent me an UberX rather than trying to force me into taking UberPool.

If you’re not familiar with UberPool, it’s Uber’s ride-sharing feature. If you use it, you get asked where you’re going, then it pairs you up with one or two other people to ride somewhere together. This costs less than other Uber ride options, but it means waiting for other people to get picked up and dropped off, which you might not want to do. UberPool was introduced in San Francisco in 2014 and became available in Washington, DC, in October 2015. But it only started barging its way into the workflow of Uber apps in DC about a month ago, in late April. San Francisco and New York are still safe, but here in DC, the once-straightforward Uber app now pushes you into carpooling, by default.

Here’s how Uber used to work in DC and still works in San Francisco and New York: open your Uber app, tap Set Pickup Location, maybe enter your destination (this happens in some locations but not all), and an UberX car shows up for you — often in a matter of minutes. If you wanted a different kind of car, like an uberXL to carry you and a few friends, or an UberSUV to hold your big suitcase, you could select that by sliding a tiny car icon below your choice at the bottom of the Uber app’s screen.

Now, you open the app, tap Set Pickup Location, and are immediately asked to enter your destination. So you do it, somewhat puzzled.

“How many seats do you need?” An in-app message asks innocently, suddenly displaying options to tap: 1, 2, 3, 4.

“Umm, one seat, like always, Uber,” you think to yourself, selecting the “1” from the list.

The next screen says “REQUEST POOL” at the bottom in all capital letters. This is the first reference to UberPool, but you still might not know what it is because it’s just called “POOL.” Two price options for your ride show up on this screen. The one on the left represents UberPool (but again, never uses that name) and is selected by default. It says say something like, “$13.91 share your car with 1-2 people.”

“But it’s 9:15PM, and I’m just trying to get home from a long day at the office,” you think. “I don’t want to get plopped in a car with someone or (even worse) two someones who have been pounding cocktails since happy hour began at 5PM.”

The other price option, which is not selected by default, says something like, “$18.80 & up get your own car up to four people.” This, my friends, is the UberX ride you originally wanted. But again, it’s not labeled as UberX, and it’s buried under workflow that’s designed to send you off on a carpool ride.

If you’ve made it this far in the Uber app — and a lot of people give up far before this point — you might wonder about the “& up” part in “$18.80 & up to get your own car up to four people.” How much “& up” money is this, really? This language is needed because the uberX fee is just a fare estimate, not a guaranteed fare. Along with lower costs, UberPool is a guaranteed fare, which people like.

But come on, Uber! If you’re so confident in the popularity of UberPool, let me be the one to choose that from the start. Don’t toss me into finding a carpool without my consent — especially when I just want to open the app, click twice, and get home.

A spokesperson for Uber says that the company created this two-price screen for customers who wanted to see how much UberPool cost compared to UberX. Though Uber says its UberPool numbers are doing just fine, I have a more cynical, alternative theory that there weren’t enough people taking UberPool in DC, so Uber decided to nudge people into it by default.

Of course, Uber could update or change its app at any time. In the future, the company plans to clean up some of this messy, misleading app workflow by doing a few things. It plans on taking out the capacity question, so you won’t be asked how many seats you need. And Uber admits that the screen showing two prices is confusing and that the app would be clearer if it explicitly said which service was being offered for each price. In fact, this is already done in London, where the cheaper, left-side fare is labeled UberPool and the pricier, right-side option is labeled uberX.

If you make it through all of these confusing screens and figure out how to opt for the uberX ride, the next time you use Uber, your default fare will be uberX (or UberPool, if that’s what you used).

Uber has a lot of loyal passengers who’ve created close relationships with its app. But any app that tries to force people into doing something — especially sharing a small space with a few other strangers — isn’t creating a relationship of trust. In fact, Uber’s latest app interface has encouraged me to cheat on it with smelly, old taxi cabs.

A lot of people who suffer from allergies hide inside their houses — especially in the spring. But your indoor air could be up to five times more polluted, according to the Environmental Protection Agency. To solve that problem, a company called Molekule is launching the first molecular air purifier — a two-foot-tall cylinder that sucks in and destroys the stuff that pollutes your air.

Molekule has a new technology called Photo Electrochemical Oxidation, or PECO, which uses UV-A wavelength LED lights to simulate solar light. This creates a chemical reaction on the filter’s surface. The filter takes things like bacteria, allergens, mold, chemicals, and viruses; breaks them down into elements that aren’t harmful; then releases these elements back into the air. For example, Molekule can convert 3.9 million E. Coli bacteria into trace amounts of water, carbon dioxide, and elemental nitrogen in two minutes, the company says.

This air filter can be preordered today for $499 with a year of free filters. It’s expected to ship in early 2017, and will cost $799. New filters for Molekule can be bought through a subscription for $99 a year, and the device will automatically determine when your filters need to be replaced, sending you new ones.

Of course, air filters aren’t new. HEPA filters, which are used today in many popular air filters, have been around since the Manhattan Project in the 1940s. These suck in air and capture particles. But the co-founders of Molekule say their device gets rid of pollutants that are 1,000 times smaller than what HEPA filters catch, and HEPA filters can be harmful. “HEPA filters can only catch some pollutants and can be a place where mold and bacteria multiply,” says Dilip Goswami, Molekule’s CEO and co-founder. “PECO breaks down the full spectrum of pollutants,” he says. Molekule is based on over 20 years of solar research by Dr. Yogi Goswami, Dilip’s father. Dilip co-founded the company with his father and sister, Jaya Goswami Rao.

In addition to being backed by traditional venture capital funding, Molekule was funded at the research stage by a grant from the Department of Defense, which has good reasons to be interested in a device that eliminates viruses from the air. The company is also backed by grants from the EPA. The science behind this air filter has been independently tested by the University of Minnesota and the University of South Florida.

Any new parent will tell you that scheduling is a huge aspect of life with a baby. Moms who breastfeed are especially aware of timing so they know when their baby’s next feeding is coming up — or when they’ll have to use their breast pump. Today a company called Lansinoh announced it’s bringing out the first Bluetooth-connected breast pump, which could help track pumping sessions.

The $199 Lansinoh Smartpump (available next week starting at Babies R Us) is a double-electric pump that pairs with your phone via Bluetooth, sending the date, time, and duration of each pumping session to an app on your iPhone or Android phone. Moms can also use this Lansinoh Baby app, which runs on iOS or Android, to note each time the baby is fed or has a diaper change, but plenty of other apps already offer to track and organize this data. And though the Lansinoh Smartpump auto-tracks the timing of each pumping session, moms still need to manually enter the amount of milk that they pump.

Later this year, Naya Health is expected to introduce a Smart Bottle with built-in sensors to track the amount of milk in each bottle. These bottles will then send that data back to an app on your phone. Naya Health got a lot of attention from parents last month when it announced the first hydraulic breast pump, planned for this fall. Breast pumps have historically been pneumatic, or air-based. Naya Health’s co-founder and CEO, Janica Alvarez, says that a water-based pump simulates the real feeling of a nursing baby. “It’s more like nursing a baby than using a machine,” she says. This hydraulic breast pump, called the Smart Pump, will cost $599.

You lock your front door with an app, you dim your lights using voice commands, and you adjust the temperature of your house from your phone. But is your smart home really that smart if you still pluck your eyebrows by looking in a mirror that isn’t wirelessly communicating with three other devices?

I’ve been using Simplehuman’s $250 Sensor Mirror Pro for the last week, and it’s available now on the company’s website. It’s an 8-inch round “smart” mirror that magnifies your face by 5x and 10x as you do things like apply makeup, trim pesky nose hairs and, yes, pluck your eyebrows. The mirror’s built-in sensor knows when your face approaches and turns on the mirror’s outer-rim light, which glows in over 50,000 color variations.

So yes, this mirror is geared toward a certain person who cares a lot about applying makeup in the right lighting — and doesn’t mind paying for luxury stuff. And no, it doesn’t display any data on its surface like some other ideas we’ve seen lately. But it’s part of a broader trend: connected-home technology breathing new life into otherwise dull devices. There are a lot of clever features in this mirror that make it more convenient and functional than it would be otherwise. When companies use smart home tech in the right way, products like this one suddenly become more talented.

For starters, this Simplehuman mirror uses Wi-Fi and Bluetooth to connect to an app and, from there, you can adjust its lighting to match preset light tones like restaurant, sunlight, or overcast. If you want to get more specific, take a selfie somewhere — like in your office — save it to the Simplehuman app, and your mirror will replicate that office lighting so when you apply makeup at home, it will look the same as it does under those glaring overheads.

You can link your Simplehuman mirror to your Nest Cam (via the Simplehuman app) so that your mirror can match the lighting of wherever your camera is aimed. If you have a Nest Cam trained on the front door for security, you could use its settings to apply makeup in light that matches that day’s overcast or sunny skies.

And Simplehuman’s Sensor Mirror Pro works with the ringmaster of the smart home, the Amazon Echo and its Alexa app. This lets you control the mirror with your voice, saying things like, “Alexa, ask sensor mirror to use the office setting,” or “Alexa, ask Sensor Mirror to turn the color temperature warmer.” This is especially helpful if you don’t have your phone handy and want to control the mirror. It’s also easier to ask Alexa to adjust the mirror when your fingers are covered in makeup or moisturizer and you don’t want to muck up your phone.

This mirror also integrates with IFTTT (If This Then That), so you can set up rules like getting an email reminder to charge the mirror when it’s running low on battery. You could use IFTTT to completely skip the Nest Cam step and set a rule that automatically adjusts the mirror to the overcast lighting preset on a cloudy day. You can even use IFTTT to link your mirror with Google Calendar so that the mirror auto-adjusts to its restaurant lighting when you have a dinner reservation.

Still, who cares this much about putting makeup on every morning? I sure don’t. I barely have have time to put makeup on at all, and when I do, it’s usually an edited version of what I might put on if I had more time. For people like me, one feature in this mirror is especially useful: a built-in timer / alarm that triggers the mirror’s light to blink when time is up. I used this a lot, turning it on with my voice by saying, “Alexa, ask Sensor Mirror to set timer to 10 minutes.” This saves you from glancing away from the mirror to look at a watch or phone.

I found this mirror’s Amazon Echo / Alexa integration to be a huge convenience. But not everyone will find it as easy to use with Alexa if they don’t have their mirror in the same general area as the Amazon Echo. My Echo is in my bedroom and I used the mirror in an attached bathroom, which is just a short distance away.

Another clever feature in this mirror is a small, circular mirror that magnetically sticks in the center of the Sensor Mirror Pro when you need to see something magnified by 10x. When you’re done with it, pull it off and stick it to a small hole on the back of the mirror.

The Sensor Mirror Pro skips a lot of the hassles you face with other smart home products. It’s wireless most of the time (a built-in rechargeable battery lasts four weeks) and it sits on your countertop so you don’t have to drill holes in the wall or run ugly cables through the house. It also just works when it’s not using “smart” features like Alexa voice prompts or light-matching with a Nest Cam.

It’s not a huge surprise that Simplehuman added digital smarts to a mirror. The company launched in 2000 with the goal of designing a better trash can, but soon grew to add other things around the home like shower caddies and soap pumps. Its lineup now includes squeegees, laundry hampers, and toilet plungers. These cost significantly more than, say, something you’d find at Target. Yet, as tempting as it is to roll your eyes at at $200 trash can or a $60 soap pump, the higher quality in these devices makes more sense when you think about how hard you use them each day. I own both a Simplehuman trash can (albeit a $40 one) and an a touch-free sensor soap pump (a wedding gift five years ago), and I really like using them. Simplehuman’s app already works with trash cans (to order you more trash can liners) and soap pumps (to order you more soap).

As of this week, Simplehuman’s $400 Sensor Mirror Pro Wide-view also connects to the Simplehuman app, Nest Cam, IFTTT, and Alexa for lighting adjustments. This mirror is a much bigger, three-pane deal that doesn’t have the 5x and 10x magnification of the round Sensor Mirror Pro I used. Its price and size are turn-offs for me, but someone with a larger bathroom and deep pockets might really like it.

Nearly everything in the home seems to be a candidate for newly-added smart home features, and some of them are actually working well, making old products feel new again. The value-add happens when something helps us simplify processes in our everyday lives. Just don’t hold your breath for the mirror that makes eyebrow tweezing pain-free.