Author: Steve Wildstrom

The following column is one of the last our partner Steve Wildstrom wrote for us. As many of you know, he passed away last week after losing his battle with brain cancer. However, even while he was fighting for his life and having to deal with our healthcare system, he was still using this experience to bring to the fore some of the key issues he saw regarding patient security, the problem with closed systems, the potential virtue of open systems and applauds Apple’s HealthKit program. I know how hard it was for him to write this in June but, as you can see, his mind was still very clear and his writing top notch. With this column, we honor the contribution he made to us and our Tech.pinions audience and he will always be part of the Tech.pinions team.

Tim Bajarin
President, Creative Strategies
Co-founder Tech.pinions

For those who have been following me (mostly on Facebook) you know I have had a lot of experience with the health care system. In addition to what has generally been very good treatment, I have learned about the impact of the internet and other communications on medicine. There’s a lot being done but much more is needed.

One of the pieces of news is the protection of patient security is very good. In the fairly infrequent cases such as the hacking of Anthem this winter, the attack on information has involved almost entirely financial, not medical records.

The reason is straightforward. The Health Insurance Portability and Affordability Act (HIPAA), in effect since 1996, imposes both very tough rules and stiff penalties for violation. A leak of information in the hospital is likely to produce seriously discouraging prosecution. Make the cost expensive for those who want to pass info on and you discourage trying.

But the HIPAA effort is far from free. The big problem is it is one of the major policies that gets in the way of effective sharing of information among doctors, medical centers, and hospitals. Electronic medical services, from general companies such as IBM and GE to specialists such as Epic and Cerner, should keep track of whatever is going on, from record keeping by doctors during exams to blood tests to pharmacy data to MRI and CT tests, and just about everything else.[pullquote]One piece of news is relatively younger doctors seem much more comfortable with the technology than their older colleagues.[/pullquote]

Of course, how well this is all used depends on how good the administrators are in making it useful. One piece of news is relatively younger doctors seem much more comfortable with the technology than their older colleagues. Another is the effective use in facilities both hospitals and centralized medical centers. I am a patient of Kaiser Permanente’s staff-based HMI and their national complete usage of the Epic system works well (except for occasional network problems).

Closed systems. These are, however, effectively closed systems, whether relatively small or a massive operation like Kaiser. Johns Hopkins Hospital in Baltimore has a close relationship with Kaiser Permanente Mid-Atlantic and, while both use Epic systems, don’t expect them to communicate with each other electronically. Giving Johns Hopkins copies of MRIs made by Kaiser required me to pick up DVD copies of the tests and deliver them to Baltimore. Written records need to be printed and delivered too.

That’s just a nuisance. Even worse is the situation if you get ill while traveling. If you need medical treatment while traveling, it can be very difficult for the personnel to get your records while complying with HIPAA. It’s safe, sort of–no one else will get it. But it can be ridiculously hard to transfer. Very solid systems have been built, in many cases on 20-year-old technology. There are secure and better systems out there. And, while the government’s caution is valuable, it is really time to develop better systems.

Fortunately, there some real developments. HIPAA makes make it hard–still a mixed challenge–but we are starting to get serious use of the Internet of Things medical devices to deliver important information to medical professionals. Most of these systems begin with the transformation of data on private networks, such as Qualcomm’s 2net Platform and Verizon’s Converged Health Management.

Open systems. As happens in the rest of the IoT world however, we are moving toward more open, more general, and more easily shared systems. Apple, which has a deep interest going back at least to Steve Jobs’ serious illness, has seen the value of iOS devices for the collection of information. And, in a new move for Apple, but in line with techniques strongly favored by developers and researchers, the new software is to be built using open systems.

Apple’s ResearchKit, launched this spring, has made progress in several medical apps: asthma, Parkinson’s disease, diabetes, breast cancer, and cardiovascular disease. Based on iPhone apps with Apple Watch likely to join, they use the device’s ability both to collect and transmit (of course, working within solid security features that are HIPAA requirements) information from the patient to researchers.

For example, Share the Journey, organized by Sage Bionetworks, is for women being treated for breast cancer. They regularly enter information on activity and perform specific tasks to improve researchers’ experience on what is happening in the case, allowing them to study and improve treatment.

Health patients. Stanford’s MyHeart Counts uses the iPhone itself to collect and transmit a considerable variety of information on what’s happening in the body. Unlike other ResearchKit efforts designed primarily for the treatment of conditions, MyHeart Counts wants to look at the data generated by healthy patients to gather information.

It’s not surprising that the ResearchKit efforts are focused on some of the dominant conditions. It’s a likely place for relatively quick gains. The provisions that make the sharing of information through public but secure networks offers an even greater advantage.

I don’t know if Apple finds much business out of this, but it is a big and important deal. To the extent they can push for better communication of health care information of all sorts while maintaining security, it is a even better gain.

CVS is a huge brand with 7,600 drug retailers whose specialty seemed to have been disorganized stores that were out of stock on whatever it was I needed. IBM is something quite different; an old-line professional company focused on enterprises willing to pay a lot for high-end services. Their partnership seems unlikely — a combination for improved medical service.

It’s clear both companies needed to do something. Drug stores need action about markets to compete with companies such as Amazon and any other online services to deprive existing retailers of profits. One direction stores have been adding is medical services, from flu shots to basic treatment. IBM meanwhile, has faced an army of increased competition in services and shrinking profits.

But IBM has an incentive of its own to build a better service. It has gone through a movement of traditional IBM design to build ways to develop large, open-system databases that can be searched rapidly for answers to questions. The company cleverly showed the public the phenomenon on national television by using a machine called Watson. It starred in the 2011 TV quiz show Jeopardy!, where it defeated human champions Ken Jennings and Brad Rutter.

For IBM, it only looked like a stunt. But the real goal was to make Watson a powerful search tool for discovery of information in the world. It has taken a while before the business really took off, but health care has turned out to be a leading choice. Improved information is badly needed in medicine, where there is a lot of information often poorly organized, but there is research effort and a lot of money. It’s no accident that Watson’s first commercial effort was a search of lung cancer data for Memorial Sloan-Kettering. The IBM-CVS move, which oddly has not yet been given a name, has launched a big new approach.

Although the financial arrangement between IBM and CVS has not been disclosed, it is clear how both sides can be winners. CVS gets access to a wealth of information from Watson that will improve its ability to diagnose and treat customers. Ultimately, this capability could help CVS to eventually move its business from cosmetics and pharmacy goods to the quick exams and shots it handles today, to true medical centers.

IBM’s gain is different. Its goal is to use Watson to collect, sort, and classify the tons of information it can take in through the CVS program. The advantage of Watson is the more data it gathers, the more knowledge it has about the total situation.

Kyu Rhee, IBM’s chief medical officer, is counting on major progress:

Today, we’re taking another step forward with a sweeping partnership with CVS Health aimed at transforming the way individuals and their caregivers engage with essential members of that patient-centered, community-based primary care team—CVS pharmacists and healthcare providers at your local CVS MinuteClinic.  These team members are committed to not only address your acute healthcare needs including a sore throat or a sinus infection, but also manage key chronic conditions including heart disease, diabetes, asthma, arthritis, depression, and cancer.

IBM is planning to use information it collects to combine medical and pharmacy records, wearables, fitness devices, home monitoring devices, and mobile apps.

Another goal is to use the CVS-based services to help carry out doctors’ instructions without needing more appointments. CVS and Watson together can help keep track of physicians’ plans.

IBM is also bringing a growing collection of newly acquired services to the effort. Phytel has built its own integrated population health management software. Explorys is a platform for sampling big data information across the platform. And Cúram Software creates social program management organizations. The goal, according to IBM Watson executive Mike Rhodin is, “extraordinary capabilities in patient engagement and population health, enabling us to deliver a powerful solution that supports individuals wherever they are, whenever they need it.”

Of course, the effort is not without risks. A particularly big one is protecting the massive medical information that IBM and CVS will be collecting and transferring between each other. CVS already has a less than good record of protecting customer records. It has lost a number of credit card records, though there has not been too much medical information stolen. We cannot even tell what IBM is collecting with serviced because today, if data is taken, it’s going to be blamed on other companies.

It’s also critical that patient information gathered by CVS and IBM be passed on to third parties, such as pharmacy companies to address discovered customer needs. Keeping anything involving the relationship between CVS and IBM private is vital and any action must be passed on to the patient, not a third company.

The problem is the information being exchanged between IBM and CVS is a lot more of a risk than mere credit card numbers. Protecting actual medical information must not only provide the privacy that patients expect but the extensive requirements of the HIPAA law and other government and insurance company terms. And, of course, we do not yet know how much it will cost CVS, IBM, or customers.

All of which probably explains why it will take a while for the service to become broadly available. But the agreement is likely to produce a major new medical service. That could be big deal.

I love Twitter. I hate Twitter. Twitter has become, for me, more important than even email, a regular report for the things going on in the world. It is also overwhelmed by junk that makes reading most difficult the iPhone.

When there is real news breaking in the world, Twitter produces the most important information. And through the day, you get news, entertainment, education, and more. But there is still a big problem — it produces a lot of valuable information but provides very little for the service to make a profit.

I’ve been a Twitter member for seven years and have watched it move from silly reports (“I just got up”) to much more useful information. Unfortunately, there is also a lot more junk every day, much of it endless advertising. And far too many posts by people using much worse style to create what they say.

The real key is Twitter on the iPhone (or, secondarily Android). It’s actually at its best on the iPad but a tablet or, even worse, a laptop, isn’t the best choice for the feed. The images are a lot better on the iPad but it is kind of hard to keep it in your pocket whenever you want it. The phone is often just too small for what is displayed.

I have used a number of apps, but mainly Tweetbot and Twitter’s own effort. The recent experience of Tweetbot, long my favorite, is trying so many different changes that it has become almost impossible to use. Between the ads thrown in by the app, the ads thrown in by comment posters, and ads with references to vendors, they create a flood of junk you don’t want.

For example, this is a typical Twitter display. Do I care about CucumberTony? Not really. And on Tweeter, you can expect every three or four such screens to include an ad like this. Add the number of individual posts and it gets worse.

One of the worst problems with Tweetbot posts on iPhone is inserts. The problems are actually all over the place but the worst feature is inserts that simply do not fit because they are too small to read and cannot be enlarged adequately.

Consider this one, not one of the worst from Re/code. The basic display is too small:

Click once to show the display on this example and you get the example, which is still unreadable on a phone:

And if you click a second time, you get an image that is much bigger and harder to read. You can slide across the screen but it can’t be read. And you cannot rotate the phone 90 degrees, which would make it readable.

There are many other problems. Just about all of the frequent submissions by Forbes are unreadable because of the announcement, ads, and other additions the publisher has crammed in that leave articles unreadable.

Some of the behavior of iPhone images is terrible. When you touch the screen in certain places–its similar but different on Tweetbot and Tweeter–the iPhone or other phone features block the screen until you tap again. (touch features should improve this in the upcoming iOS 9.)

So far the trend looks to get worse. I have no desire to raise money from anything I post. But that is not true for many other posts who are desperate for earnings. And the harder they try, from both posters and the app owners, the worse it gets for readers.

I can’t give up because it means too much to me. But it sure is getting worse.

Many of us have dealt with prerecorded treatment voices at medical centers. And many of us have found it frustrating, and worse if the situation seems bad, before you get to talk to a skilled human being. But the situation is getting a lot better as the virtual “professionals” take advantage of rapidly increasing action to provide both faster treatment and more efficient, lower cost.

One of the most interesting is Sense.ly, a new service built on top of the code designs offered by Expect Lab’s MindMeld service. The goal is to build highly accurate, thorough voice services on top of data, language understanding, machine learning, continuous processing, and user interface over devices.

We’re ready for help. “There’s a lot of need for action,” says Sense.ly CEO Adam Odessy. “One problem is frequency. We need to communicate and limit the amount of direct contact. We focus on long-term process, Natural language is key to developing to virtual nurse. The overall goal is to communicate more and reduce cost. In the end, we always need a nurse or doctor, but we can eliminate a lot of cost.”

Sense.ly is built around a virtual nurse named Molly–somehow a bit cuter than necessary and with just enough machine-generated language you are certain it isn’t a real human being on the other end. The initial plan is to set up systems for three areas of medicine–medical institutions, including hospitals and other clinic setups; insurance companies, and pharmacy companies.

They all stand to profit, although their roles are different. Medical institutions, obviously, are designed to get treatment to patients and will certainly seem more useful for consumers. Insurance companies hope to collect large amounts of information that shows them what is going on in the field, designed to distribute service–and to lower the costs. The pharmacy is interested in collecting information, often in the form of inquiries from current patients, to monitor, gather information, and distribute drug use.

The effective use is both the cleverness of the human-like robot helper at the clinic with increased ability to collect the most useful information, especially when it can be collected using mobile devices to provide information about patients. A goal is to make it possible for the service to constantly remember data, especially useful with the ill and aging patients.

Three services demonstrate some plans that combine information provided by devices and discussions with Molly:

Congestive heart failure solution: The practice is designed to collect critical information and deliver close CHF information to the caregivers based on devices and conversations. Data tests include daily blood sugar and blood pressure, diet monitoring, and weight tracking.

Diabetes solution: The most effective way to treat diabetes is to collect daily information to keep close track of glucose levels, designed to both improve diabetes itself and lower the risk of heart conditions. The roles of patients and devices include daily glucose measurement, blood pressure, ocular examination, and podiatrist health.

Behavior health solution. This gets a little out of the normal treatment field. The goal is end-to-end behavioral treatment for patients, including automatic assessments defined by clinicians. Steps include reports of daily symptoms, daily photos of patients, compliance monitoring, and regular guidance manuals.

The goal is to use as many technologies as possible, working through phones, tablets, and PCs. The hope is in making Molly’s avatar role as human as possible, with  speech-recognition conversation as normal as possible. Using data collected by patients and by wired or wireless devices is critical. And video and augmented reality sent to patients can help in treatment.

A number of large players are involved in Sense.ly already including the UK National Health Service, University of California San Francisco, Kaiser Permanente, Universidad Nacional Autónoma de Mexico,  Novartis, Allscripts, Panasonic, and Microsoft.

And, of course, the efforts must comply closely with the rules imposed by the government. Everything is built around Health Insurance Portability and Accountability Act requirements and Sense.ly has been working the HIPAA requirements for the past year and a half of development to qualify.

A step is likely to assist the development of programs built on MindMeld and Sense.ly, as well as competitors in the works, is likely to take continuing service from ever more capable devices, phones and watches as well as specialized equipment for IoT services.

Apple is not a part of the Sense.ly project but the company is using the new HealthKit software to help bring in roles for iPhones and Apple Watches. “HealthKit is combining integration,” says Sense.ly’s Odessky. “It provides additional vital signs and better data. We look forward to everything from everything from cameras to observer stuff like FitBit to get more precise information.”

The use of health services from IoT based data studies to devices with ever more accurate information of patients’ medical services is a big deal and it is one of the best ideas for remote devices. Better and more efficient could be a real winner for everyone.

In a recent column, Bob O’Donnell took a look at the rapid growth of the Internet of Things. The IoT is a large variable collection, from heartbeat monitors on your wrist to a camera watching your garage. It can work with apps, send info to friends or web sites, and control events in your house. It is important, but the discussion of IoT for consumers is only part of the story.

For corporations where the IoT effort has been going on longer, it is already making a major difference. Certain critical industries, like oil and gas production, electrical utilities, etc., have a critical need for tracking events and being warned of goings on at work sites. Sensitive detectors in the systems have been part of IoT for a long time. But the traditional method of communication and control has been SCADA, designed specifically for running large operations. The system worked by getting information into a central controller and, in many cases, security depended mainly on pure complexity.

IoT was a natural approach that is catching on. Like consumer IoT systems, items in the field are equipped with lots of abilities. The devices have a set of communications capabilities and they are set up to exchange information with both mobile devices and other communications on the network.

In business systems, devices can send key messages directly to sites and provide better and simpler communications while also sending information along to headquarters for analytics. The internet can provide both lower cost and improved communications better than the old POTS lines did. And well delivered techniques can be added to provide acceptable levels of security.

Of course, the users of IoT systems have very different needs. Retailing, for example, wants faster information from the stores, Panasonic’s Powershelf (pictured) being an example of such a system. Printed product labels on shelves are starting to be replaced by electronic devices. Of course, they use an LED display to repeat the label information. But their communication capability is used to send reports of the purchases to warehouses so they can maintain stock and to headquarters so they can get on-time information on sales in all the stores. And, of course, electronically updated prices can be sent to both the electronic label and the point of sale. You could even use the information of IoT exchanges to change prices regularly in the course of a day. With IoT, retailers get tricks that once worked only on the internet.

The development of IoT is changing the nature of how corporations use data. Databases have frequently been used to analyze corporate information, but in the traditional way of collection; that is, feeding in the data and getting back reports. But now the data is often a lot greater and flows a lot faster. One more has been to move from traditional databases to new ways of moving through methods such as Hadoop.

Another is a change enterprise is learning from competitors. Oddly enough, modern web systems designed by startup companies are now being copied by large corporations for their own IoT. Traditional old-fashioned monolithic computing setups are being replaced by microservice systems, in which only a single computer (more likely than not, a virtual machine) holds a single piece of software. Each of these computer “containers” talks to each other with standard language to share their capability. The next step is to get this company system onto a cloud, whether the company’s own private setup or a commercial cloud. Amazon Web Service is by far the leader but the move of traditional companies is causing the old types such as IBM to push the business hard.

As has been the case in consumer computing, IoT is not the right approach for everything for companies. Many systems, such as the standard databases and the maintaining of taxes and payroll, are going to continue. But IoT is bringing a lot of the access to other sorts of information.

In many cases, bolder use of IoT by customers is generally feeding into to the corporate level. The same systems, from airlines to grocery stores, is a way to provide IoT to customers. Over time, a growing increase in information flows into the company. It is a lot of trouble for airlines to keep track of their flight, staff, pricing, and other phenomena to keep the systems working—and IoT can bring dramatic change.

A RIP announcement often gives us word of a celebrity who has passed away after they have left the public mind. The death of AOL, with its purchase by Verizon for $4.4 billion, is certainly the end of a company that is no longer what it once was. But AOL was a very big story and one worth recovery.

The fact is America Online, as it was known in the early days, was responsible for the birth of consumer communications. It was born in 1989 out of some even earlier services as a dial up–in those days, the only way to think about it–information service. Back then, the internet was government controlled and mostly limited to academics. Other services, such as CompuServe and MCI Mail, were interested mainly in businesses.

Steve Case had a very different idea. This was to be a service for millions of consumers. In its early years, it added dozens of sources of information and invited members to discussion groups. (Back then it was only text; neither audio nor video existed on dial up networks and even pictures were a challenge.) AOL lined up members by flooding the U.S. with sign up software on floppy disks, often thrown in with magazines and newspapers.

AOL was clearly a success. It did in competitors, such as Prodigy, a company created by CBS, IBM, and Sears. It invented all sorts of online conversations. Years before Reddit came up with Ask Me Anything sessions on the internet, magazines put their stories online and journalists on AOL to discuss content (I did it fairly regularly for BusinessWeek).

Email, which had been a service in academia and some businesses, became popular for the public as “You’ve got Mail” let you know about an incoming announcement. (And a movie in 1998, a remake of the better film The Shop Around the Corner.)

AOL made itself a leader in the internet business. It bought Netscape, the leading web browser and the builder of a considerable net service. Interestingly, the purchase of Netscape was $4.4 billion, just a bit (without adjustment for inflation) more than the Verizon-AOL deal.

But AOL’s secret was the company’s business still stood on dial-in calls to the service. As 2000 approached, services such as DSL and TV cable began replacing dial up. AOL remained a sign up opportunity for customers who had other connections, but there were also lots of other services, like Microsoft’s MSN and others. AOL could go on sending out sign ups, now on CDs, but the field was getting less popular.

Services were also being attacked. Once upon a time, AOL messages were for content and conversation. We even used them for quick and easy chats at work. But they were eventually overcome by SMS, not to mention services from Apple iMessage to WhatsApp.

People managed not to notice the internet group that was squeezing AOL. Case and Gerald Levin announced the merger of AOL and Time Warner, creating a massive combination of movies, television, books, magazines, and the internet.

It seemed like a great company, but it somehow never worked. Time Warner, already dealing with the problems of publishing and Hollywood movie-making, never joined with AOL effectively. Case, who had served as chairman, lost interest in less than five years and went off to focus on investments in the health business. AOL, which moved from its merged corporate headquarters in Manhattan to its old home near Virginia’s Dulles Airport, faded away. An acquisition has been expected for a long time.

One question about the Verizon purchase is the future of some AOL properties that really do not seem to fit, especially with an ongoing fight over net neutrality. Huffington Post, a major news publisher with a generally liberal tilt, and two leading tech sites, Engadget and TechCrunch, do not seem to fit well with Verizon. Re/code’s Kara Swisher reported talks were already in progress to sell HuffPo to various buyers, with the leader being Germany’s Axel Springer.

Whatever happens in the end, there appears to be not much left of AOL except ads and some members. The end of AOL is not much of a surprise.

The other day, I wanted to take a photo that had been sent to me attached to an email and make it a feature in a Facebook post. Now, if I had been working on a PC–Mac, Windows, or even Linux—all I would have to do is select, copy, and paste or just select it and drag.

But I was working on an iPad. You have several ways to do it that will transfer the picture. They’ll work better than they used to with earlier apps and iOS versions, but they are still unpredictable and considerably more painful than on a PC.

This difference is found in a numbers of ways, many of them much less graceful, for linking between different apps on phones or tablets. (While the particulars are different, iOS and Android offer similar issues. I’m sticking to the Apple experience because I have more experience.) That is why I spend a lot of my time on a phone or an iPad to follow events. I will generally switch to a PC when I face an option for working with text, images, or numbers. While you can do almost all of the same tricks, especially on an iPad with a keyboard, they are a lot easier on a PC.[1]

The reason is not that Mac and Windows designs are smart while tablets and phones are dumb. The fact is both approaches are very smart at what they do. The important difference are what they do and how.

The basic operating systems in use today are based on developments of the 1990s. A critical feature is the operating systems are built around a big file system. A file system app, File Explorer in Windows and the Finder on a Mac, lets you find any file stored on the system. Use Notepad or Text Editor and you can open many files that make no sense to a human. If you foolishly mess around with a critical XML file that is part of an application, it may really damage the code (most apps these days, like Word for either Windows or Mac, will stop you from loading files that make no sense). The important fact is every program can see whatever is stored on the machines and use what it needs.

The code of phones and tablets is very different. Android and iPhone have file structures, of course, but their software keeps users away from their discovery. On Android, you can add an app that gives some access to file listing; on an iOS devices you cannot do that with any third party app you can get through the App Store. As Apple says it in the File System Development Code, “The iOS file system is geared toward apps running on their own. To keep the system simple, users of iOS devices do not have direct access to the file system and apps are expected to follow this convention.” [pullquote]Android and iPhone have file structures, of course, but their software keeps users away from the discovery. On Android, you can add an app that gives some access to file listing; on an iOS devices you cannot do that with any third party app you can get through the App Store.[/pullquote]

Over time and with some improvement in phone software, ways have been added to move content among some apps. For example, a finger point in an image, or using an icon of a square with an arrow pointing up will let you move a copy of the picture to a number of specific alternative apps, including Message, Mail, Facebook, Twitter, and Flicker. But if you want to move a photo to a Word document, you will just end up with the file name inserted into your document instead of the image. If you really want the picture, you have to get it into Photos and then select it from Photos within Word. Given the complexities, if you’re lucky you’ll manage to memorize transfer rules that work for a few apps. And editing or resizing the pictures turns out to be more trouble than it is worth.

That’s why I find it hard being away from a PC for long. Say I’m writing a column for Tech.pinions on a PC. I’ll write my text in the WordPress HTML editing setup, sometimes I’ll write in another editor. If I do that, it’s easy to move what I have written, usually through a straight copy-and-paste. I’ll select an image from a variety of sources and often move to Photoshop, more likely than not just to alter the shape and size. Finding web references and adding to the piece is far easier on a PC than an iPad or phone, partly because it is easy to move the HTML address but even more because it is easy to have more than a single window open on the screen.

The point isn’t that you can’t pull it off on an iPad or a phone; it is that the project ends up being a lot more work than shifting to a PC. In some cases, the difference is even more extreme. I don’t like looking at spreadsheets on a 12” or 13” laptop, let alone a tablet or a phone. You can do a quick, simple video on a phone, but editing a video of any length and quality requires a PC.

Don’t get me wrong, I love both a smartphone and an iPhone (I haven’t been convinced that other tablets really are desirable tools). I find I use a PC a lot less than I used to. But there remains a lot of tasks best done on a PC. It is why, though the PCs share of the market has declined, it’s not going away.

[1] I am not including the Microsoft Surface and other Windows tablet-like systems. They can be tablet-like hardware, but the software is Windows and, for good or bad, they function like PCs.