Paragent.com

Refreshing Desktop Management

Paragent Offers Open Source Computer Management Software

without comments

Paragent computer management software is perhaps the most effective way to manage your network while keeping costs down. Paragent is open source software and while it is free to download and use, it offers an extremely easy to use interface and a wide variety of premium features.

Some of Paragent’s computer management software features that are popular with the many organizations include comprehensive desktop management, hardware and software asset inventory, real time alerts for IT support staff when issues occur, integrated help desk, seamless remote desktop and advanced search functionality. Paragent is known within the industry for offering top rated software as a service solution. In addition, being open source software, Paragent computer management software is incredibly flexible and affordable to implement.

Paragent computer management software is available for free download, but it should be know that for organizations big and small, Paragent offers premium support for pennies per day. In fact, with their top tiered support package, your organization can receive high quality IT support for less than 70 cents per month per computer.

For more information regarding Paragent computer management software or to download the open source code, please visit our main website at the following link: http://www.paragent.com

Written by admin

February 13th, 2008 at 6:04 pm

Posted in Uncategorized

Lisp and Agile Development

without comments

We were contacted by a reporter at InfoQ. He saw the “Why Lisp?” blog post, and wanted to ask about how Lisp helped us with any Agile development practices. You can read the resulting article here. We don’t tend to follow a strict Agile methodology at Paragent, but instead have worked to incorporate some of the Agile practices into our day-to-day routine.

One important aspect of Agile practices is the idea of rapid release cycles. Instead of slaving away for months at a time, and then dumping a huge release in your customer’s lap, Agile practices promote much shorter cycles - on the order of weeks. The basic idea is to keep the development pipeline as short as possible. If you mess up something at the beginning of a development cycle, and only find out six months later, it is very expensive to flush the pipeline. If you get back to them two weeks later, you can recover from your mistakes quickly, correct, and move on.

There is a corollary with the internal development process called the Edit-Compile-Test loop. It is the time it takes between writing code in your editor, and finding out whether it works or not. With traditional programming environments, and large applications, this cycle can be extremely long. This tends to result in programmers doing more in the edit phase of the cycle, to amortize the relatively constant-cost compile phase of the cycle. This has impacts across many aspects of development. For example, if you do more in the edit cycle, it means that your commits into the revision control system will be much larger, and cover potentially disparate areas of code, making rollbacks and merges more difficult.

I am a bad programmer. I have very bad habits. When working in C++, I would bang away on code sometimes for days without fully compiling, testing or checking in. A 15 minute compile is such a buzz-kill. When I was in the flow, I didn’t want anything to break my stride. Programming in Lisp (and from what I understand languages like Ruby, although I have what amounts to no experience with it) is a completely different experience. The compile cost of bringing up an application happens when you first start it, and unless your application is a true behemoth, it is certainly no worse than starting even a minor Visual Studio solution. At that point, the compile phase of the Edit-Compile-Test loop is negligible. You basically edit, and then test to see if it worked. If so, you can check in that little snippet and move on. It may be one of those things you just have to experience to truly appreciate. What is even more impressive is that you are not sacrificing program efficiency on the altar of programming productivity. When you hit C-c C-c in emacs, you are getting a compiled version of your function inserted into the runtime.

What is really interesting is how this Lisp work-flow influences the entire software development stack. Peter Seibel has a great chapter on integrated testing in Lisp, and there are several other good tools out there. Perhaps one of the most often quoted, but never followed piece of software development wisdom: “Premature optimization is the root of all evil.” — Donald Knuth. Lisp makes it dead-easy. I’ve profiled orders of magnitude more Lisp code than I have in any other language just because I could. Lisp is more than just a bunch of parenthesis thrown together back in the 50’s before we could afford lowercase letters and called a language. Lisp brings a different attitude to programming.

Don’t get me wrong - I’m still a bad programmer, with many bad habits. I make optimistic estimates of how long code will take to write. I check in code I haven’t fully tested. I don’t comment my code well. I start coding before I have necessarily thought out why, how, or where I am going. I’m a bad, bad person. Lisp makes it a little bit easier for me to be good some times.

Written by tritchey

November 7th, 2007 at 11:36 am

Posted in Lisp

Why Lisp?

without comments

One advantage to a proprietary hosted application is that no one cares what you use to get the job done, as long as it works. Now that we are open source, there is understandable interest in what that source code looks like. In our case, it turns out the majority of our software is written in a language called Lisp. For most people that even recognize the name, Lisp is that funky artificial intelligence language with all the parenthesis that was cool in the 80s, but is now dead.

Lisp is not dead, it just smells funnyEdi Weitz

So, why did we choose Lisp? Part of the answer involves Paragent’s history. When Paragent started, our goal was to create a development platform that would allow programmers to take advantage of idle computing resources in a corporation. We developed a peer-to-peer agent that spoke a special xml-based language that would make it easy to write programs and distribute the workload across many machines. When we decided to shift our focus from a development platform to a hosted desktop management service, we took a good hard look at what we had. It became pretty clear that we were a poster child for Greenspun’s Tenth Rule:

Any sufficiently complicated C or Fortran program contains an ad hoc, informally-specified, bug-ridden, slow implementation of half of Common Lisp.

We decided to take a closer look at Lisp as a possible implementation language for the new server component, along with Ruby/Rails, PHP and Python. I won’t go into detail on why Lisp is so great as there are many resources available that do a much better job than I can. Perhaps one of the most influential in much of Lisp’s recent resurgence is Paul Graham’s Beating the Averages. In addition, a new book by Peter Seibel called “Practical Common Lisp” had been released around the same time that we were considering implementation alternatives. He has a wonderful chapter on why lisp, and then followed it up with 31 more chapters of great Lisp programming education. After much deliberation, we decided to give Lisp a go.

True to the promises, we were able to develop and launch the first version of Paragent in record time with a limited budget. We were adding minor features every week, and then came out six months later with major new functionality including remote desktop and trouble tickets. In addition to the development productivity afforded by Lisp, we benefited in areas such as maintenance and support. One example is the ability to update running code on live production servers without missing a beat, sometimes with customers literally looking over our shoulders! Those are just a few of the obvious benefits from using Lisp, but there are hundreds of small ways that Lisp makes our jobs more fun every day.

Hopefully this piques your interest a little in this funny old AI language with all the parenthesis. If you want to learn more, check out the links above, download the Paragent code, and join the community!

Written by tritchey

October 21st, 2007 at 11:34 pm

Posted in open source

New Installer

without comments

We have uploaded a new installation file that should make it much easier to get the Paragent platform up and running on your own system. It is a tar.gz file, and includes a script for installing on the Fedora distribution. You can get the file at the Paragent google project site here.

Written by tritchey

October 15th, 2007 at 1:33 pm

Posted in open source

Why Open Source SaaS?

without comments

The primary method used by most web 2.0 companies to gain mindshare is to give away a “free” version of the service, and then hope they can up-sell a portion of those users to a premium service. The general rule of thumb in the industry right now is that for every 100 free customers, 1 will pay you money for your service. 100,000 users, and 1,000 will pay you money. Often, the up-sell is based on some value-add. More storage, additional features, no ads. At this point, it is almost formulaic.

Unfortunately the Paragent service doesn’t work that way. These aren’t 100,000 users, 80% of which will use the service for a month and never come back. For the Paragent application, behind each user are all the computers they are monitoring. And whether the user ever comes back to the site or not, that is tens, or potentially hundreds, of persistent SSL connections that we have to maintain. 100,000 users could be 2,000,000 potential agents connecting to us if we averaged 20 computers per account. We could limit the number of computers a free account could monitor, but the basic value is in the automation. Limit the number of computers you can monitor and maintain, and the service loses any value whatsoever. 90% of the heavy lifting in Paragent is simply maintaining the connection to the computer. The economics of giving away our service is just not the same.

That is the technical reason that the traditional web 2.0 freemium model doesn’t work for Paragent. There is also a more practical reason: a closed source SaaS model is bad for our customers. In the unlikely event that Microsoft closes its doors today, if you have Microsoft Word installed on your computer at this moment, you still have Word installed on your computer tomorrow. What happens if SalesForce were to close down, or 37signals? What happens if you can’t log into your sales force automation tool when you wake up one morning? It doesn’t take an extremely unlikely event to expose customers to the risk of closed source hosted service providers either. Acquisition is the number one goal of almost every single Web 2.0 company out there today, whether they admit it or not. A company could simply sell out, and the new owner decide they don’t like the current business model. Witness the firestorm after the flickr sale to Yahoo over a simple change in login accounts.

SaaS and closed source is the worst of both worlds. We believe our hosted model and services are key to our value proposition. By releasing our code as open source, we are giving our customers the long-term stability and flexibility of knowing they will always have options.

Written by tritchey

October 11th, 2007 at 5:12 pm

Posted in open source

Paragent is Open Source

without comments

While it won’t be news to anyone that has gotten to this blog from our website, I wanted to write an entry announcing the fact that Paragent is now an open source company. 

We have been working on this move for a while now, and there is still a lot more work to get done.  It is not unlike getting a phone call that the in-laws are coming over for a visit. We are taking a look over the house and making sure all the dirty socks are in the laundry basket, and the pizza boxes are in the trash. We have created a new web site, and are hosting the project at code.google.com. Part of the process of moving from a closed-source company to open source is going over all the code that we are releasing. This means adding copyright/licensing boilerplate; scrubbing for passwords, emails and any other Paragent-specific details; creating documentation for community members; and bringing our current customers up to speed on what this all means. 

It turns out that story is pretty easy: Paragent.com is still focused on providing a safe, secure hosted service that provides a cost-effective and headache-free tool for IT administrators to help organize their lives. The value we were providing yesterday is still the value we are providing today. However, if you are a developer, or an IT shop that likes getting under the hood, Paragent is now available as an open source download. We are even here to help with our expert and professional team to get you started, and keep you running. 

In future blog posts, I will go into more detail about the hows and whys of moving to an open source model. Subscribe to the feed, check out the new site content, and download the code!

Written by tritchey

October 3rd, 2007 at 4:22 pm

Posted in open source

Tagged with