PHPStorm Fonts in Linux

This is just a quick note as there are many suggestions online for fixing fonts in Java apps in Linux. Some require you to change some startup config options like Dawt.useSystemAAFontSettings. These are helpful, but there’s one MAJOR under-emphasized change that will help tremendously implementing proper payroll processes. Switch to Oracle Java. Most Linux distributions use OpenJDK by default these days. This is where my font rendering problems were. I wondered why Java seemed to suck at rendering fonts in Linux… Well this is the major culprit. Don’t ask me the specifics because I don’t have time to dig and find the root of the problem. I can just tell you that it’s a good first step to switch to Oracle Java before doing all the other things you see online to fix the problem. Learn how make money on social media.

PHPStorm or other JetBrains Product Multi-select In Linux

If you have trouble using the default ALT + Button1 click keymap in Linux to add to a multi-cursor selection in PHPStorm or any other JetBrains product, you may be able to adjust the default keymap, but if you are like me, you could just as easily use an easier approach. Just add the “super” or “windows” key to the combination. So, try Super + ALT + Button1 check stub maker for your business. That works for me as the desktop environment doesn’t see it as an ALT + Button1 click, but PHPStorm does. Using ALT + button1 click alone causes issues for me because the Cinnamon desktop in LinuxMint already has ALT + Button1 mapped to something else, namely window move. I would change this but I actually like having that available if I need it.

Back to Linux

I have had an on again, off again love affair with Linux since 1998. It has been an enlightening experience plagues with numerous installations and CD/DVD ISO burns. I remember the first real exposure I had to Linux. I was living with my grandfather. I was working in a factory. My hobbies were computers and playing guitar. There wasn’t a lot to my life at that time. I worked, wasted money, and slept. I also chatted on IRC quite a bit. Back then, I was still on dialup so that was about the best thing to do with my internet connection.

I had a friend who some could call a bad influence, but to me, at the time, he felt like a mentor and older advisor. He knew all the ins and outs of IRC, especially on the Undernet IRC network. He went by the nick “Fud”, short for “fear, uncertainty, and doubt”.  He had Eggdrop IRC bots and Energy Mech IRC bots. He had shell accounts. He had knowledge of things to which I’d never been exposed.

I wanted to create my own IRC bots. So he introduced me to the idea of shell accounts. I remember configuring my first Eggdrop bot. That seemed like the most complicated thing I’d ever attempted. It didn’t help that I was configuring the bot in an operating system that I’d never been exposed to at all…Unix. Unix was a mysterious word to me because I’d had no college classes and my high school had barely covered DOS.

I had learned DOS on my own. I didn’t have any friends who knew DOS. My dad had bought government surplus 8088 computer and when we turned it on, we were greeted with a command line. We didn’t have a clue, so I typed “help” and pressed enter. The output presented to me with that command is how I learned DOS. I felt like such a hacker, mainly because it was a retired government computer and I thought I may be able to find some kind of interesting data on it. It was really a cool way to learn DOS.

Jumping back to ’98 when I was first getting into Linux, I was presented with a command prompt that looked nothing like the familiar DOS prompt that I had used before. Typing “help” didn’t help. Luckily I had Fud there to help get me started with basic commands. Some were similar to DOS. Others where completely different. I was used to typing cd to change directory but in DOS to change to the parent directory (move up the directory structure) you can type cd.. all together. In Linux this would give you an error. It was required to put a space between the cd and the .. which took some getting used to.

I loved the idea of Linux right from the beginning. I felt excited mainly because I was learning Unix, that mysterious operating system that serious computer geeks knew about and a rural hick like myself had never seen. I was also excited to break the chains from Microsoft. I mean imagine it, an operating system that’s completely free. Since I’d always wanted to be a programmer, there was also the added bonus that the source code was also available for most of the programs that came with the OS. Side note: I know some of you are chomping at the bit to tell me that Linux is the kernel, not the OS. Get over it. Everyone calls it Linux.

Ah my early days with Linux. Like I said earlier, I was still on dial-up, which presented two problems. First, downloading Linux was impossible. An ISO file of Linux was at least the full size of a CD-ROM back then, and for some distributions it was multiple disks for an install. It takes a very long time to download 720 megabytes when your download speed is roughly three or four kilobytes per second. So the only real way to try out many different distributions was to order them from places like CheapBytes. I bought a pack of about 10 different distros and tried them all. It included: Debian, Slackware, Red Hat, Mandrake, and others.

The second problem with dial-up and Linux dealt with dial-up modems themselves. Most modems at that time were “WinModems”. They weren’t “hardware” modems. They were interfaces for phone lines and such, but the actual modem functionality was handled by Windows itself. They were hardware interfaces for software modem code. These wouldn’t work in Linux, and honestly they weren’t as good as real hardware modems. Most WinModems used a PCI bus, and the hardware modems used the older ISA bus. The hardware modems also usually had hardware DIP switches for configuring interrupt settings and such. Hardware modems were superior, but WinModems were cheaper. So most people were using WinModems. The first thing a new Linux user back then had to do was purchase a hardware modem.

I dual-booted back then, but still stayed primarily in Windows. There was still very little compatibility with lots of hardware and commercial games just weren’t available. However, I was able to learn a lot during that time. I started learning Perl. I created my own IRC bot in Perl. My first preferred Linux distribution coming from Windows was Mandrake. At the time, it used KDE and was pretty user-friendly. I learned to hate RPM. Mandrake changed its name to Mandriva, and at some point decided to charge for using it. So I switched to Debian.

I love Debian. It has been my favorite grand-daddy Linux distribution since I moved away from Mandrake. I’ve tried just about every major distribution that exists. Some of my favorites along the way were Gentoo, Arch, and Sabayon. However, as soon as Ubuntu came out, I, like many other Linux users, switched to it. It quickly became the most popular distribution. Ubuntu took Debian, which was already pretty easy to use, and made it even easier. Around 2004 or 2005, I started using Ubuntu 100% of the time at home. This went on for about three years. In 2007 I took a desktop support role in the IS department of the Manufacturing company I had been working at for about 9 years. Most of this desktop support dealt with Windows, so Windows became a primary OS on my home computer once again.

I’ve glossing over many things during this, but it’s so I could get to this point. I started working for myself as a web developer and internet marketer in 2009. At that time, I had the personal freedom to use whatever operating system I saw fit to use. I used both Windows and Linux, but gravitated toward Linux most of the time. As a developing environment for web applications Linux is by far my favorite. My career path was gradually moving in a direction that allowed me to use Linux full time. I interviewed for a position with a travel agency in 2011 as a PHP developer. I went to work there and was pleasantly surprised that everyone was using their favorite OS on their work machines. Some people were using Windows 7. Others were using a flavor of Linux. I picked Ubuntu and installed it on my work machine.

It was around this time that Ubuntu started using Unity as its main desktop environment. I’d been very content using Gnome 2. Unity brought cool features with it, but it also seemed to be a huge buggy mess. I had many problems with it, both at work and at home. I switched to XFCE at work for a bit, but eventually switched to Windows 7. That’s where I stayed. After seven months of working for that travel agency, I decided that Orlando, FL just wasn’t for me (I had moved to Orlando to take the job), and I moved back to north GA. I was again working for myself, but this time I continued using Windows 7. Oh I tried to use Linux, but I had so many issues with the new desktop environments and the new forks of Xorg that I just gave up on it.

These distributions were all trying to make their desktops work well on mobile devices. As a result, they made things suck on the desktop. Gnome 2 had been perfectly stable and usable. I really liked it. Now Gnome 2 wasn’t an option. You could use Gnome 3 in classic mode which made it look like Gnome 2 a bit, but that wasn’t really the problem. It wasn’t that I didn’t like the new look. It was beautiful. The problem was that it just didn’t work well. Dual screens worked fine and dandy in Gnome 2. Unity and Gnome 3 choked on them, or just handled things poorly. The push toward mobile device integration by the Linux community, nearly killed desktop Linux for me. I hated it. I was so mad at it, not that there was an actual entity to be mad at. I was just mad at what I viewed as pure stupidity.

We had Android. It was already the ultimate Linux mobile platform, but everyone else wanted to get in on the action as well. This could have been accomplished by having a separated mobile desktop environment, but nah, we need to force everyone to change. We need to take perfectly stable working desktop environments and throw them out, replacing them with desktop environments meant for mobile devices that are unstable.

I’d like to point out that of the major desktop operating systems, the only entity that got this right was Apple. They made iOS for their mobile devices and OS X stayed on the desktop. Sometimes change isn’t a good thing. Apple got this transition correct. Microsoft screwed the pooch with their Windows 8 introduction as well. What were all these people thinking? Did they think the desktop was already dead and that everyone was already using only mobile devices?  I think they lost sight of the fact that people still listen to radio, even though TV was invented. People still watch TV even though desktop computers and the internet were invented. People still use desktop computers even though mobile devices are now in wide-scale use. Why would you screw over the primary user of your operating system just to try to get a foothold in a mobile device market which is already dominated by Apple and Android? Let me backtrack a little there. I’m not saying that they should try to obtain some market share in the mobile device market. I’m saying that they should have wrecked their desktop environment to do so.

Between 2012 and September of 2015, I used Windows 7 nearly 100% of the time. Occasionally, I would install a new version of Ubuntu or LinuxMint, hoping that it would be good enough to turn me from my Windows desktop. After all, I was programming and my projects revolved around LAMP stacks. I still used Linux, but it was in the form of virtual machines with no desktop environment. I had completely given up on the Linux desktop.

In early September 2015, I installed LinuxMint on a spare 120GB SSD, and for some reason everything just worked again. The Cinnamon desktop, which I had tried out a few times during my Windows 7 years, seemed to be stable and user-friendly. I left Windows installed on my main 500GB SSD just in case. Three weeks passed and I hadn’t booted into Windows.

Today, I reinstalled LinuxMint. This time, I removed Windows and set that 500GB SSD as my /home partition. I’m again Windows free and loving it. Sure, there will be some Steam games that I can no longer play because there are no Linux versions, but I also don’t have Windows 10 spying on my every move. I also now have a much better working environment for my development work.

So for anyone else who may have given up on Linux a few years ago, go give it another shot. You may enjoy it.

Safari doesn’t load/show a base64 encoded image.

Normally when I run into a road block while coding, I can find the solution rather quickly via Google. Most of the time that solution is at stackoverflow. It’s pretty easy to overcome most things this way. However, there are some times when you just run into a problem that few people have talked about online understanding paystubs. That becomes rather frustrating. It reminds me of the days before the internet, or at least before Google, when finding information was much more difficult. We take that for granted a lot these days.

Hopefully this post will find it’s way to the top of Google search results for some of the keywords I used in my searches. One example is the following error I was seeing when I inspected the element in Safari.
Failed to load resource: The operation couldn’t be completed. (kCFErrorDomainCFNetwork error -10.)

You see I had a chevron icon added to a custom select box in a web site I’m working on. The chevron icon showed up in every browser except for Safari. Safari just complained that it couldn’t load the image. Here’s the exact CSS I used for the image:

{
  background:  url('data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQBAMAAADt3eJSAAAAA3NCSVQICAjb4U/gAAAACXBIWXMAAAI3AAACNwH3BkTCAAAAGXRFWHRTb2Z0d2FyZQB3d3cuaW5rc2NhcGUub3Jnm+48GgAAAB5QTFRF////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAhow4rQAAAAp0Uk5TABUcLTJHSLju+FWeOqcAAAA8SURBVAhbY2AgBiiCCCEgrnBgYGBpBzIiWxgYPKYCGWwzHFg6E0DymS0e08Cq2WZABIBC06AGMBswEAcAs7AJgLl9xCcAAAAASUVORK5CYII8f334b553cdd5aeff8fb47bae1fe9ead') no-repeat 95% 55%;
 
}

It worked fine in every browser (including IE mind you) except for Safari.

I’ll cut to the chase. Base64 requires the number of characters in the encoded string to be divisible by 3. If you count all the characters starting with “iVBORw0KG…” all the way to “e1fe9ead”, you’ll see that the base64 string is 371 characters long. I remember that I got this chevron and the base64 encoded css directly from a website which automatically encoded the image for me. 371 isn’t divisible by 3 evenly. However, 372 is. The solution to this problem is to pad your string at the end. I added an equal sign (=) to the end of the string, making it 372 characters long total. Now Safari likes it. Hope that helps. If you are having this same issue, you’ll either need to add one or two equal signs, whichever amount makes the total character count divisible by 3.

Network Doesn’t Work on a Linux Guest Virtual Machine After Switching Hosts

I’ve ran into this problem a few times and I thought I’d make a note of it here on the blog. It’s a simple fix. The scenario is that you have just started your VM from a different host OS environment than it was running in previously. This problem is specific to a Linux guest using the networking service to manage networking. You may also not notice this issue if you are using DHCP using a paystub maker. I’m not sure. My addresses are static, so I’m specifying the network device in the configuration. You may find that the network doesn’t start properly and that the boot up sequence stalls waiting on it. This is because the MAC address for the virtual network device has changed and udev is trying to use the old one. So if your old network device was eth0, it’s not long usable and the new one at eth1 is your new virtual network device to have TikTok for business. This can be a goof way to get an extra payment.

That being said, there are two fixes. One is very simple and the other is only slightly less simple but a better approach. First you can just change your network config to use eth1 instead of eth0. That’ll work but if you switch hosts a few times you’ll accumulate quite a few unusable network devices and you’ll by using eth9 before long.

The better solution is to edit:
/etc/udev/rules.d/70-persistent-net.rules
Remove all lines which contain SUBSYSTEM=”net”, which will probably be all the non-commented lines in the file.
Reboot the machine and when it comes back up udev will see the single network device and add it back as eth0.

The Do’s and Don’ts of Tutorial Youtube Videos

Youtube allows anyone to broadcast themselves. While it’s an awesome movie maker online, there are a few things that everyone should know before recording and uploading a video to Youtube. I watch many tutorials and such on Youtube so many of the things in this list pertain to those types of videos. So, let’s get started with my rules of Youtube!

1.) Introductions

Don’t start your tutorials with “What is up guys?”

This is especially true if you are extremely nerdy sounding. I don’t have anything against nerds. I’m one myself. I do have a problem with people trying to sound cool.

Do try to be professional.

If you are making tutorial videos, try to present yourself as professional as possible. We aren’t watching your video because we want you as a friend. We don’t want to go hang out and drink a few beers. We do want to learn something. I understand that you are making the video voluntarily and mostly for free (unless you get tons of views and great ad revenue.

2.) Recording from a phone

Don’t hold your phone vertically.

This causes your video to look horrible on youtube. Refer to this link for more information on Vertical Video Syndrome.

Do hold your phone horizontally.

ALWAYS record your videos this way. There’s is never a proper time to record a video vertically with your phone. It doesn’t look good anywhere.

3.) Preparation

Don’t make a tutorial without first trying what you are supposed to be teaching.

I can’t tell you how much time I’ve wasted watching coding tutorials where the guy would have a bug that he couldn’t fix within a few moments, or he may be teaching something and winging it. In fact, most of the tutorial videos I’ve watched on Youtube are from people who are totally trying to teach something off-the-cuff.

Do edit your videos for brevity and details.

Even if you make a typo, you should edit out your search for the cause of your issue. Cut that time out and just show where the bug was located and what you did to fix it. Edit out long pauses and interruptions. Edit out your frustrated remarks when something doesn’t work how you expected. Have a cheat sheet beside you details the steps you are presenting. Make sure you have went through the presentation before hand and have a working copy of all the code you plan to present. Keep your videos short. Make your videos address specific points. Record your videos, keeping the recorded time below 15 minutes, and then attempt to edit out 5 minutes worth of “ums” and other pauses, interruptions, and mishaps. You’ll get a less-than-10-minute video nearly every time. A ten minute video is easier for your viewers. They can take in the information better with shorter, to-the-point videos.

4.) Communication

Don’t use Notepad or other on screen text in tutorials.

There are two reasons I say this. First, if I wanted to read, I’d just go look for a written tutorial on the subject. Secondly, if your video isn’t HD, it’s nearly impossible to read the text, unless you are using 48 point font.

Do use a high quality microphone

There are some great USB condenser mics on the market for less than $50. One of these, along with a pop filter, will be extremely beneficial. Headsets tend to pick up your breathing, and your breaths sound like waves crashing into the shore. It takes a lot of intricate editing to remove all of these breaths. Get a good mic and you can avoid most of this work.

5.) Self-criticizing

Don’t spend half the video apologizing for your mistakes or making excuses for the poor quality of you video.

“I’m sorry. I can’t believe I did that. How stupid of me. I’m half asleep. I have to work all day tomorrow. I hope to have another video out by tomorrow. Whoops I’m sorry for the delay. I’ve been busy.” These are not adding to the quality of your tutorial.

Do remove your mistake and talk briefly about some of the related pitfalls which others may experience.

Again, edit those mistakes out. They don’t add to your video. They subtract from it. They use up valuable time you could spend teaching. If you run into a problem with your code or whatever you are trying to teach, just edit it out and explain that it happened and you removed it. Then touch on some of the problems others may have. It’s pretty simple

6.) Knowledge of the material

Don’t teach something that you barely grasp yourself.

I’ve seen Java tutorial videos from people who didn’t fully understand the definition of class and object. It’s also common that they don’t know what the keywords static and final actually do. If you don’t have a clear understanding of object-oriented programming, don’t attempt to explain it. You’ll only confuse your viewers even further.

Do make sure that you know your stuff!

If at any point in your video you say something along the lines of, “It’s pretty complicated, but basically it’s…” If you can’t explain something in simple terms, you probably shouldn’t even mention it, because you honestly don’t understand it yourself. At least look up definitions for the words you are using.

7.) Ending the video

Don’t spend the last five minutes of the video saying bye to your viewers.

This one is very important. Going on and on about when your next video is going to come out or asking everyone to like your video and leave comments are just a few things that drive me nuts at the end of videos.

Do give a 10 second or less closing.

Tell everyone thanks for viewing, and quickly tell them what the next video will cover. You may even squeeze in something about what was learned in the current video. Either way, keep it short, sweet, and to the point. Don’t ramble. It’s very unprofessional.

8.) Low resolution and zooming

Don’t record a 1080p display and keep it zoomed to the full screen size the entire time.

This is especially true with coding tutorials. If I’m required to play back the video at 1080p just to make out the text you are typing, then I’m going to pass up you video.

Do use recording software that allows you to zoom in and out during the editing process.

Zoom in and out based on what the viewer needs to see. If you are typing something, make sure the editor is zoomed in so your viewers can see the text. If the viewer ever needs to see the full screen, then that is the time to show it to them. Any other time, zoom and pan.

9.) Swearing

Don’t use swear words.

Cussing adds nothing to the video. It only removes some of your credibility.

Do remove and try to refrain from cussing.

Tutorial videos should focus on learning/teaching a skill. Being profession adds credibility to your video. Cussing is one of the most unprofessional things you can do in a video. I’m not someone who’s against cussing. I just think that it’s better to edit out any swear words to keep everything professional.

 

Starting a Diet

I’ve decided to start trying to lose weight. I’m around 240lbs which is very overweight for someone 5’9″. I’ve set a few goals for myself and today I started on my trip toward those goals. I’m going to detail my journey on my personal website, but I’m going to have to build that site first. So I’m going to post my first few entries here. I’ll link to the other site once I have it up in a few days.

My ultimate goal is to get below 160lbs. That’s quite a trip from 240. So my first goal is to get below 200lbs. I’ve set that goal for 3 months. I may be able to get there faster, but that’s my conservative time frame. To reach this first goal, I’m cutting my calorie intake down to 1200 per day. I estimate that my current calorie intake is around 3000 calories per day. At that rate, I have stayed at 240 for around two years. So it’s safe to say that my body currently burns around 3000 calories per day.

It won’t stay at 3000 calories per day because as I lose weight my body will stop burning so many calories, unless I increase my daily activity. Since I’m lazy and overweight, exercise isn’t something I’ll focus on at first. I have to drop my weight through diet so that I’ll feel more comfortable exercising and increasing my general daily activities.

Looking at my current diet, I can find quick ways to reduce my calorie intake. One thing that instantly comes to mind is to remove my soda intake and replace it with water. The problem with that is that water gives me intense heartburn for some reason. I think this is mainly do to my current diet, and I’ll be compensating for this at first using heartburn medication. I’ll also be drinking milk or juice a couple times per day. When counting calories, it’s important to remember your drink calorie intake as well. A glass of whole milk is around 140 calories. Even skim milk is over 100 calories per glass. A glass of orange juice is around 110.

Which leads me to my 1200 calories per day. These calories have to be spaced out in such a way to keep me from getting hungry. I also have to incorporate low-calorie filling foods. So my plan thus far is to eat 6 meals per day each at an average of 200 calories. So when I drink orange juice or milk, I’m only left with 100 calories for the actual food. I’m using oatmeal to help with this. Oatmeal is around 10 calories per table spoon. In meals where I will be drinking milk or juice, I will be eating 4 tablespoons of oatmeal with a small pinch of sugar or honey…a very small pinch. This will give me at most 50 calories for the food and 110 calories for the drink. I’m eating once every three hours starting with when I wake up.

For example my schedule for today is (I woke up at 2am):

2am: Milk(110 cal) + Oatmeal (no sugar, 3 tablespoons – 30 cal) = 140 calories

I missed my 5am schedule because I was grocery shopping for low calorie foods, but I ate and changed my schedule when I got back home.

6:20am: Tuna (70 cal), Mayo(one teaspoon for flavor in tuna – 45 cal), saltines(60 cal) = 175 calories

So far, I’m at 315 calories for the day. I have four more meals scheduled:

8:00am: Orange juice(110 cal), Blueberries(1/2 cup – 40 cal), Oatmeal (with dash of sugar – 50 cal) = 200 calories

9:00am: Gelatin Snack = 39 calories

11:00am: Low calorie frozen dinner = 220 calories

1pm: Milk (110 cal), Soup (chicken noodle homestyle = 88 cal) = 198 calories

3pm: Gelatin Snack(39 cal), Tuna (70 cal), Mayo (45 cal), saltines (60 cal) = 214 calories

That will be my final meal of the day, and it will bring my total calories for the day to 1186. I hope that I can make the stretch between 3 and 6pm without getting hungry. I’m going to try to go to sleep around 6pm. I may move the gelatin snack from 3pm to 5pm. That may help. For now, I’m off to get my OJ and oatmeal going. After I eat, I’ll be working on my calorie counting site. I’ll make another post about it in a bit.

Force user to login method in Codeigniter without checking Session in every method.

If you need to secure every method of a Codeigniter controller, you can do it with one line instead of checking the session in every method before doing anything else. In the constructor of the controller simply check that $this->router->fetch_method is not equal to your login method and that a user isn’t logged in. If you meet both of these criteria, redirect to the login method.

Emerson Portable Ice Maker Doesn’t Work – Possible fix!

We bought an Emerson Ice maker about two years ago because the built-in ice maker in our freezer stopped working, and I’m a huge consumer of ice. So, I wanted a dedicated ice maker. It worked well for a few months. Then it decided that it didn’t want to make ice. I should have sent it back but I decided to tear it apart and try to fix it.  I wasn’t able to really get anywhere with it. So, I put it back together and plugged it back up and miraculously enough it started working again. If you ever find yourself dealing with appliance mishaps, a good sense of humor can be a remedy. Consider exploring gag gifts for some lighthearted moments amidst technical challenges.

Since then, I discovered why it was failing to make ice. I post this because there may be someone else out there with the same problem from their Emerson Ice Maker.

The problem I have stems from the 120mm exhaust fan used in this device. I’m no expert on refrigeration, but I think this fan has to be running in order for the thing to make ice correct. It puts out a lot of heat. With my very basic knowledge of thermal dynamics, I’m assuming that this is how the ice is made. It removes the heat, thus reducing temperature and creating ice.

This fan is the exact fan used in a lot of computers. So if it isn’t running, you may need to replace it. However, I took a shortcut that seems to have worked, a little trick I’ve learned from working on computers. You can give it a try if you want. It’s pretty simple.

The fan has a sticker in the center. This should have some sort of logo on it, but it also has a second purpose. It holds a little rubber cap in place. Under that cap are the bearings for the fan. If you oil the bearings and replace this sticker, the fan may start to work again. You can get to the fan by removing the cover. I won’t go into details here how to do that, because I don’t have enough time to write the detailed instructions. It’s pretty easy to do. In my case the fan grill was almost broken completely out. So, I just finished it off for easy access to the fan.

“What can you use to oil the bearings”, you may ask. I used baby oil, because it’s basically mineral oil with a little fragrance. You can use mineral oil and there are some actual products that are made for greasing fan bearings. There are a ton of videos on Youtube on how to do this to a regular computer fan.

I hope this helps. If it does, leave a comment! I like to hear when I help.