Some quick thoughts about Raygun.io

Raygun.io has been on my personal radar of technology to check out since seeing their shiny t-shirts at a conference last year (they do kind of stand out, in a good way). So I grabbed a trial account to take a look, figuring I’d write about what I found.

Firstly, what is Raygun? In brief, it’s real time error management that’s available for a range of platforms, designed to be a “smarter” system. If that’s not descriptive enough, then there’s more information on their website.

Initially my thinking was that I was going to be a technical post, however it became clear pretty quickly that there really wasn’t going to be a large amount of value in a technical post – because it “just works” ™. It’s quick and easy to implement, and there’s really not a lot more that I’d be adding on top of the existing documentation (I only used the .NET and JavaScript providers, however the trial on offer means you’d have the opportunity to test any of the providers before making a commitment). There are also other posts popping up covering the technical aspects, such as this one, which is slightly specific to Windows Mobile but still covers other aspects of the product.

However not needing to focus on the technical aspects is a positive thing, because the opportunities that Raygun presents around error related workflow and potential process improvements are where the full value of the product lies. It’d be easy to simply place Raygun in the category of “just another logging provider”, however the fact is that it is much more than that. Sure there’s a logging provider component to the offering (and the fact that one of those is a provider for JavaScript is an attractive bonus), however the real value is in the combination of the providers and the portal, and the processes that you can build around the pairing of the two.

The portal is simple but powerful, with clean styling. The opening dashboard contains an overview of the latest errors in each of the categories (Active, Resolved, Ignored, Permanently Ignored) with information displayed in ways that imply the involvement of a UX specialist. Translated: Shiny Web 3.0 (we’re not doing 2.0 anymore, right?) stylings and shiny graphs!

Identical errors are grouped together so rather than seeing a huge list of repeating text you can focus on the actual error and see at a glance the number of times it has occurred. This might seem like a small thing, but with the high signal to noise ratio one typically encounters with error management it has powerful results, and allows you to easily focus on the most pressing issues.

The Activity Log shows comments and status changes (workflow related) of an error, i.e. marking it as resolved and any comments that your team might choose to make. As you’d expect, if an error that has been previously marked as resolved reoccurs then Raygun sets its status back to active and notifies you. This allows you to focus on working with all occurrences of an error, rather than on a single instance. That’s a pretty subtle but important shift from other options for straight error logging.

The portal lets you use permalinks for specific errors so you can bookmark them and save time. This combined well with mobile device support (that’s continually being improved) to keep an eye on things if/when you’re on the move.

You can define multiple applications on an account, and can set email notification options on a per-application basis. Raygun can email you when an error occurs and/or you can choose to receive a daily digest of errors. There are also (optional) options and features for team management, so Raygun is suitable for small to large situations.

That’s just some of what’s available, and there’s more information on the website, specifically on the features page.

Raygun is a commercial app so you’re going to want to look at the pricing sheet and pick a plan, however you can easily justify the cost by looking at the amount of time you or your team would take to build something similar. Chances are you’ve worked somewhere that’s built their own error management portal. Some are better than others (I’ve seen quite a few), but I’d wager that even the most simple of attempts would work out to be more costly (and less feature laden) than Raygun when you calculate the cost of building and maintaining them. YMMV, but that’s my opinion.

Raygun is a Mindscape product, so anyone who’s used LightSpeed (or any of their other products) probably knows what that means – listening to customers, rapid iterations and frequent addition of new features/bug fixes. In fact there’s been quite a few iterations and changes while this post has been in draft (improved mobile support on the portal, improved management of company accounts, and improvements to the summary emails to name a few that I remember offhand).

In summary I really like Raygyun, and while we’ve not yet fully implemented it into our own internal products we do plan to do so (time permitting!), and we have already recommended and implemented it for a customer where it immediately offered quick wins through increased visibility of errors. Raygun is not just a set of logging providers, but more it’s a tool that enables you to implement some powerful processes around your error management. At the very least it should make you stop and think about how you can be more proactive with your errors, rather than just sweeping them under the rug.

Tags: Mindscape, Web Development

posted @ Wednesday, June 19, 2013 6:46 PM | Feedback (0)

Insert Generic Cloud Based Title Here

A while back, I came across this Mac centric post titled This Chart Proves Microsoft’s Worst Nightmare Has Come True. If you don’t feel like clicking the link then I’ll save you the trouble – it’s a chart showing that PC growth rates are rapidly declining, and that you can (somewhat unsurprisingly) place the launch of the iPad on the chart’s timeline to see that it had some affect on things.

However, I’m not sure that this is much of a nightmare for Microsoft. Sure, their lives would have been easier if nothing had ever changed, but I’m pretty sure they saw this coming. What makes me think that? This: Microsoft announces Windows Azure is now used by over 50% of Fortune 500 companies.

I’m no analyst, but to me it seems like Microsoft timed their move into cloud pretty well.

Tags: Azure, Microsoft

posted @ Sunday, June 16, 2013 3:58 PM | Feedback (0)

Azure and Visual Studio integration–it’s the little things

While upgrading an Azure project to target version 2.0 of the SDK, I got curious as to how things might look for anyone who didn’t have the new Azure tools installed. For this project, the team is incredibly geographically distributed (5 countries and counting), so I thought I’d fire up another machine and take a peek.

I don’t know what I was expecting, but for some reason I wasn’t expecting it to be handled as cleanly as this:

It’s the little things like this that make developing in Visual Studio so pleasant. It’s also why it’s hard to go past Azure for all things cloudy if you’re already working in the Microsoft space, as the level of integration really is hard to beat.

Tags: Visual Studio, Windows Azure

posted @ Thursday, May 23, 2013 11:25 AM | Feedback (0)

Linked in Endorsing isn’t working… or is it?

Like many people, I don’t really use Linkedin much at all. However I do keep my profile reasonably up to date - so when they added the “Skills and Expertise” area along with a system for endorsements I added some items to my profile and left it at that.

A few months down the track, I really don’t think the system is working. Some of the endorsements I’ve received have been from people who have never seen any evidence of my knowledge in the skill they’re endorsing me for. These endorsements are from people I haven’t seen for over 10 years, when my life and career definitely had a different focus, and when some of the things they’re endorsing me for didn’t even exist. I’ll take the endorsements, as in fact I *do* happen to know about the skills I listed on my profile, however the point is that there’s absolutely no way to verify any of this despite the “social” nature of the system, which is where we get to a couple of the big issues I have with Linkedin.

First is the spammy and “in your face” nature of a lot of Linkedin’s reminders and notifications. They’re constantly emailing, trying to convert me to a premium account (No! I don’t want to HURRY to claim your offer of a free month of Linkedin Premium! You’ve been offering me that for 6 months!), and this spamcentric approach applies to the way they handle endorsement notifications, although these are less about email and more about taking up a chunk of screen real estate. When the big section appears with the big “ENDORSE ALL” button, it’s going to be quicker for some people just to click the thing and get rid of it.

Secondly is the “you endorse me for something and I’ll endorse you for something” back scratching culture that Linkedin seems to cultivate (whether they mean to or not). That was certainly how it felt that many recommendations occurred, and it seems like skills endorsements have gone a similar way. I get that this is a symptom of how the modern world works, but the thought that every endorsement out there is just someone trying to improve their own profile is a sad thing, and adds to the reduced value of Linkedin as a service.

So the outcome is a spammy social media site, where a large part of the value added crowd sourced network graph based information about other people comes from a desire to make message boxes go away as quickly as possible, or a selfish need to have endorsements or recommendations reciprocated. That doesn’t sound too useful to me.

Disclaimer: I have “Magnets”, “Paradigm Shifts” and “Enhanced Synergistic Outcomes” listed in my Profile’s Skills and Expertise section, so you may not want to take my opinion of Linked in too seriously.

Tags: Social Networks

posted @ Tuesday, May 07, 2013 10:29 AM | Feedback (0)

Link Roundup–6th of May 2013 (Azure special)

I’ve spent the last couple of months involved with a project that makes heavy use of Windows Azure. On top of that, there’s been a lot of recent enhancements and improvements in the world of Azure, hence this post has an Azure theme, covering Azure related things which have appeared over the last few weeks.

Windows Azure: Active Directory Release, New Backup Service + Web Site Monitoring and Log Improvements – some new features released on the 8th of April - cloud based Active Directory, Windows Azure backups, and diagnostic improvements to Azure Web Sites.

Introducing the Event-Driven Message Programming Model for the Windows Azure Service Bus and What's New in the Windows Azure SDK 2.0 Release (April 2013) both cover some new features that were recently added to the Windows Azure Service Bus specific. The new event driven programming model looks handy. TL;DR – some cool stuff was released for the Service Bus.

Penny Pinching in the Cloud: Enabling New Relic Performance Monitoring on Windows Azure Websites – using New Relic to obtain handy performance monitoring information from your Azure based sites (TL;DR – oooh shiny graphs!).

Windows Azure SDK 2.0 for .NET Released - Some cool new stuff – improved publishing and management for Web sites (which of course are still in preview mode), streaming diagnostic logs for Web sites, updates to the Visual Studio Table Explorer, performance updates, new powershell cmdlets, and more. I’m looking forward to Web sites coming out of preview, as they’re looking more and more like a viable (and cost effective) option for some of our hosting.

If you’re someone who liked videos, then you can also learn more about the SDK release, and see some demos of it in action, from Scottgu’s visit to the Cloud Cover Show on Channel9:

Publishing to Azure Web Sites from any git/hg repo – handy deployment options for Azure Websites where the code is hosted on either:

• a local git repository
• a TFS project
• a git project on GitHub
• a git or Mercurial project on Bitbucket
• a git or Mercurial project on CodePlex
• code in a Dropbox folder

Tags: Microsoft, Windows Azure

posted @ Monday, May 06, 2013 5:11 PM | Feedback (0)

CSS Explained

posted @ Wednesday, April 03, 2013 4:16 PM | Feedback (0)

3 Potentially Useful SublimeText Tips

Here’s a few useful Sublime Text related items. I’ve been using Sublime Text as my text editor of choice for many years and don’t think I’ve even scratched the surface of what it can do yet.

JsFormat is a JavaScript formatting plugin for Sublime Text 2

Flatland is a simple theme and accompanying color scheme for Sublime Text 2 – based on the Soda theme, looks nice.

Syncing Sublime Text 2 Settings via Dropbox – contains instructions for both Windows and OSX.

Tags: Sublime Text

posted @ Wednesday, April 03, 2013 4:14 PM | Feedback (0)

Link Roundup – 19th of March 2013

Hacking the <a> tag in 100 characters – obvious but interesting. Just when you thought you’d made your Mom safer online by hovering over a hyperlink to make sure it’s going to where she thinks it’s going (although you could argue that 3498349 layers of URL shorteners over every damn thing are already ruining that for her).

Makeover Your Résumé as an Interactive Infographic – uses infogr.am which looks pretty cool.

Macbook Air Deconstruction – neat. Although at the end the lights come on, but, does it boot?

JSON debugger/visualizer for Visual Studio 2012 – In the past I’ve used Sublime Text with some JSON syntax highlighting as well as some simple web based tools so having this in the IDE is handy.

The Harlem Shake for iOS – open sourced, for all your iOS based Harlem Shaking requirements.

Export your Google Reader feeds – useful for anyone who’s never actually looked at where the export function lives! It’d be so easy to fill this post with links entirely about Google Reader shutting down.

posted @ Tuesday, March 19, 2013 8:25 AM | Feedback (0)

On VMware Fusion with genetically modified Macs

A while back when I ripped open by Macbook Pro to install a second HDD I commented on how it felt a little ‘wrong’. I also mentioned a semi random crippling speed issue with VMware Fusion and the Optibay (actually it’s related to any additional hard drive by the sounds of it).

Well it seems that the thread and the issue are both still alive and well and frustrating people well over a year after I originally created it.

I’m kind of surprised by that, but have to wonder whether it’s the sort of issue that pops up due to the closed/controlled nature of Apple hardware. Developers aren’t used to needing to cater for a huge variety of hardware configurations, and instead probably (I’m guessing, clearly) get used to simply checking what model and era they’re executing on and assuming that no evil hackery has gone on since the machine was shipped from the hallowed sanctum of Cupertino (or from China, whatever).

Granted, the nature of virtualisation (or virtualization for those who prefer the z) is as such that it’s working at a much lower level than a lot of standard software, so it’s a class of software that’s more prone to issues with non-standard hardware, but I’d be really interested in whether there’s a lot of other examples out there of software issues like this popping up with Apple hardware that’s been modified.

The issue was crippling, and made it impossible to use virtualisation, so in the end, my “fix” was to buy a new MBP. I still stand by what I said in my original post in that the MCE Optibay is a great modification, but I really did find the whole experience “interesting” and can’t see myself ever wanting to modify any Apple hardware I may or may not own in the future. I also find myself wondering, if maybe that’s not entirely unintentional on Apple’s part.

Tags: OSX, Apple

posted @ Thursday, March 07, 2013 9:43 PM | Feedback (0)

Link Roundup–16th of Feb 2013

Securing Web Application Technologies – “The SWAT Checklist provides and easy to reference set of best practices that raise awareness and help development teams create more secure applications. It's a first step toward building a base of security knowledge around web application security. Use this checklist to identify the minimum standard that is required to neutralize vulnerabilities in your critical applications.'” (or grab the poster here in PDF format)

6 Reflections After 3 Years of Business – in the style of Patrick McKenzie. Summaries like this are useful – this post inspired me to do a quick (private) one of my own, and there’s something about seeing numbers visualised that can be incredibly powerful for both planning and reflection (yes, I realise that’s stating the obvious a little).

Step by step from jQuery to Backbone – experienced in jQuery but interested in Backbone? Well, click the link.

Extending the ASP.NET Optimization Framework – a couple of useful ideas for customising ASP.NET’s bundling and minification functionality introduced in v4.5. Particularly useful for anyone using jQuery templates.

Bundling and minifying in ASP.NET MVC – a basic 101 post introducing the aforementioned functionality, with a couple of tips and tricks thrown in.

Holy Grail of eCommerce Conversion Optimization - 91 Point Checklist and Infographic – I found the infographic a bit of a mess, but when you look at the items in list form there’s a few there that are good points and reminders. There’s also a lot of recommending services that are probably affiliates of SEOMoz.

posted @ Saturday, February 16, 2013 1:31 PM | Feedback (0)