H3RALD - Tag 'software' (RSS Feed)

Book Review: Succeeding with Agile

h3rald@h3rald.com — Sun, 25 Apr 2010 12:16:28 -0000

“This is not a book for those who are completely new to Scrum or agile. There are other books, classes, and even websites for that. If you are completely new to Scrum, start with one of those.”

— Mike Cohn, Succeeding with Agile

Great. That’s just great. Good job I started with the Introduction first, otherwise the first chapters of this book would have been way too overwhelming!

Succeeding with Agile is a book that doesn’t teach you about Scrum or agile methodologies, it won’t give you a definition of ScrumMaster, sprint, or backlog… instead, it takes all that for granted and teaches how to pragmatically adopt — or better, ADAPT to — Scrum, in the context of yourself, your team, and even your entire organization.

“[…] this book draws on my experience with Scrum over the past 15 years, but especialle the last 4. For the last 4 years, every evening after I spent the day with one of my clients, I would go back to my hotel room and make notes about problems they were facing, the question they asked, and the advice I gave.”

Indeed, this book is a gold mine of information, anecdotes, tips and tricks about everything you could possibly want to know about making Scrum work, at any level. If you have some knowledge about agile development you definitely have some questions: will it work? … is it really more productive? … how can I make my boss understant this?. This book has all the answers you need. Most definitely, it also answer questions you didn’t think of.

If you don’t know what all this is about, then you’d better do your homework first:

Overview

The book is organized into five parts of different length, ranging from 20 to over 100 pages. If you read the book from the start till the very end, you’ll notice that the start of each part is like a new milestone in Scrum adoption: first the author makes sure that you are prepared (Part 1), then moves on to deal with individuals and initial resistance (Part 2), then teams (Part 3) and finally the whole organization (Part 4), until you can finally taste the fruits of you labor (Part 5).

In a way, you may well want to carry this book in your briefcase every day you go to work, and read it bit by bit, as you make progress in your quest for Scrum adoption.

Part I: Getting Started

Part I is about making sure you know why becoming gile is important and beneficial to you and your work environment. It will teach you how to promote Scrum, its advantages and challenges, and the different ways to go about it: Start Small or Go All In? Stealth or Public Display? Things like that. Pointless theory? Not really: everything is well documented, with success stories to support one way or the other.

Part II: Individuals

This part was very interesting from a psychological point of view: it deals with individuals and their possible reactions to becoming agile. You’ll meet skeptics, followers, saboteurs and diehards — no hope? Well, of course not: you’ll learn how to deal with each one of them in the best way possible. This part will also introduce you to new roles and responsabilities related to Scrum.

Part III: Teams

Up next, Teams. You’re no longer dealing with single-minded individuals, but with more complex groups. New challenges emerge, mostly related to communication and people interactions. I particularly enjoyed Chapter 13 — The Product Backlog, which provides invaluable insights on this important everyday tool. Chapter 15 — Planning is another interesting read: it teaches you a lot about planning vs. estimating, and coming to compromises to meet deadlines.

Part IV: The Organization

If you made it up to here, then you’re nearly done. You probably know most of the tricks by now, but there’s still a lot to learn. Chapter 17 — Scaling Scrum is definitely worth reading, even just for the analysis between formal and informal communities, while Chapter 19 —Cohexisting with Other Approaches almost feels heretical at times: mixing Scrum with Waterfall? Is that even conceivable? Yes. Sometimes it’s the only way, especially when you have to deal with compliance to standards like ISO9001. Once again, the author has a nice success story on how a company passed an ISO9001 audit by providing documentation in form of photocopied notes and by adding a single failing test to persuade the auditor that the automated test suite was not rigged. Priceless.

Part V: Next Steps

Only two chapters in this part of the book, which mainly deals with (self) assessment and progress analysis. Still worth a read, but you can safely leave it out for when you succeeded with agile.

Technical Analysis

I’m not exaggerating when I say that this is by far the best book I’ve read in the past few years when it comes to the way it is organized. Start by reading the table of contents: if you take each chapter out and make a bulletted list of each section you’ll end up with a handy (and free!) cheat sheet on how to promote and adopt Agile methodologies.

This doesn’t mean the book isn’t a worthwhile read, but rather that it can also be used as a reference when needed.

Formatting and Readability

From a technical writing point of view, this book is spotless. I should keep it on my desk to remind me how technical documentation should be written, except that… it’s not a technical manual of course. But the formatting and the way content is laid out can make the most skilled technical writer very jealous: there’s never a huge blob of boring text, never a series of pointless pictures: Mike Cohn (or his editors) did a terrific job composing this book.

You can start reading it from any point and it still makes sense, diagrams are simple and clear, and yet extremely useful, and so are the reference tables and spreadsheets. They never hurt, they are always in the right place, at the right time. And bold text is aptly used at the start of list items, so that even if you skim through the key concepts will still make it to your brain. Excellent.

Style and Contents

Reading this book is like listening to a seminar hold by some charismatic icon like David Allen or JoAnn Hackos: you never get bored, and you constantly learn something. Mike’s informal and conversational style is one of the main reasons why you should read this book instead of others on the subject: he is a great communicator, and he knows how to make his point across.

As an added value, Mike also uses two types of boxes throughout the book:

Things to try now — Whenever a new strategy or practice is introduced, you’ll find one of these boxes containing a bulleted list. “Commit to running the next two or three sprints without any overtime”, “Do you understand what motivates every other person on your team? If not, find out. How? Ask them.”, … these are just examples of some of the author’s reccommendations to put you in the right track.
Objection — Either actual quotes from customers and employees, or possible statements which may come out throughout the process of adopting Scrum. Things like “If the product includes less than what we’ve planned, no one will buy it”, or “My team won’t self organize; team members are too passive and look to me to lead”, … of course, what makes these objection boxes valuable is not the statement themselves, but the tips on how what to do about them. There’s not a single one left unanswered: you really feel you’re covered in any situation.

Final Thoughts

I really enjoyed this book. It took me ages to read it, not only because it’s quite long (450 pages), but also because it’s very dense of information. Another author could have made it three times longer, but I was glad Mike didn’t. I’m pretty certain I’ll keep it near me and read bits from it when I need to: it’s pretty much the Bible of Scrum adoption.

What’s wrong with it then? Not much. Perhaps the only thing I really missed was an introductory 50-page-chapter on Scrum and agile. I know this is not meant to be a book for beginners, but some basic glossary or Scrum cheat sheet would have made it accessible to an even wider audience, at virtually no cost for the author or the readers, who could have just skipped that part.

Anyhow, I give it a 9 out of 10.

JournoTwit - The best way to organize your tweets

h3rald@h3rald.com — Thu, 05 Nov 2009 15:19:17 -0000

Since I started using Twitter on a regular basis, I felt overwhelmed by the endless stream of data generated by the people I was following.

The official Twitter page quickly became inadequate to manage my tweets, so I began to search for an alternative through the myriad of Twitter clients available out there, both web and desktop based. After trying out a few desktop clients, I decided to restrict the search to web clients only: between work and home, I may use up to 4 different computers and 3 different operating system, and I really didn’t fancy the idea of keeping the same client up-to-date everywhere — even if such client existed.

Over the past months I tried dozens of different web-based Twitter clients, and narrowed the list of must-have features to the following:

The interface should be simple to use and not too cluttered.
I should be able to categorize tweets in columns (à la TweetDeck).
I should be able to know, when visiting the site, how many new tweets I have to go through.
I should be able to mark tweets as read.
It should display media files (at least pictures) inline with the tweets.
The interface should provide all the most common twitter actions like reply, retweet, follow/unfollow, shorten url, upload pictures etc. etc.

The good news is that I found at least one web-based client able to do all this: JournoTwit.

Introducing JournoTwit

JournoTwit was born – as many software projects – to scratch an itch:

Probably the last thing anyone was expecting me to do — even myself, was to create my own twitter client. However, I’ve been a little fed up with not having the features I wanted and running 5 or 6 accounts, I was getting annoyed at using several different applications just to have them logged in concurrently. […] In under 24 hours I put together a twitter client that functioned enough for me to call it my one and only. A few more days and I added in enough features that I felt it was good enough for public consumption. It is however, not perfect and I have plenty of improvements on my to do list for it.

JournoTwit’s introductory post by Andrew Spode Miller (@spode)

After months of public consumption, JournoTwit became a feature-packed Twitter client able to compete with a lot of mainstream alternatives – albeit remaining always relatively unknown to the masses. You can call it a niche Twitter client, able to satisfy a few basic needs:

The ability to manage multiple Twitter accounts at once.
The ability to categorize all incoming tweets automatically, according to the type of information within them.
The ability to keep track of unread tweets.

These three features alone were enough to make JournoTwit my one and only Twitter client. And no, it’s not only for journalists and writers.

Interface overview

After logging in, JournoTwit looks like this:

Note: I am using the edge version of JournoTwit, a sort of development snapshot with the latest features.

At the top, some more-or-less intuitive icons allow you to perform all the most common global actions:

Tweet
Manual refresh
Mark all columns as read
Add new columns
Quick search
Edit settings
Logout

Next to this global toolbar, there’s a list of links, each corresponding to a column. Clicking a link toggles the visibility of the corresponding column.

In each column, tweets are displayed in different column according to their state:

Read
Unread
Selected

You can select one tweet at a time by clicking the + icon. This toggles the tweet-specific actions:

Reply
Send a direct message
Save as favorite
Retweet
Translate

Pretty intuitive and easy to use, so far.

Default columns

When you login, you’ll notice that all your tweets are not presented in the traditional, disorganized single-column stream layout. Instead, they are sorted automatically into different columns, according to their type:

My Feed: All the tweets you sent. By default, this column is minimized.
No-Mention: All tweets containing your username without the “@”, i.e. every time someone mentions you sneakily, without sending you a reply.
Messages: All the direct messages you sent and received.
Mentions: All the tweets containing your twitter username (with “@”), such as replies to your tweets.
Statuses: All the tweets posted by people you follow that do not contain any link or cannot be categorized through other columns.
News: All the tweets posted by people you follow containing links to articles or non-multimedia web pages.
Retweets: All the retweets posted by people you follow.
Visual: All the tweets posted by people you follow containing links to pictures or videos. Where possible, media is displayed inside the tweet.
Audio: Same as above, but for audio items.
Chatter: Attempts to collect all conversations involving you or people you follow.

Surprisingly, these default columns are enough to make your Twitter experience easier and more manageable, without configure a single setting. They’re obviously not perfect: some images are not resolved automatically, for example, but it works well otherwise.

Still this may not be enough for your needs or maybe simply not the right thing. No problem: JournoTwit is extremely flexible when it comes to organizing and sorting out your tweets.

Adding new columns

All columns except for Mentions and Messages can be modified as you see fit. These two columns cannot be modified simply because there’s nothing you need to modify it, if you think about it. But they can be deleted, of course (and re-created in a blink, if you delete them by mistake).

Let’s go through the slightly geeky process of creating a column.

When you click the Add New Columns icon on the top-left corner you’ll be prompted to further clarify whether you want to add a…

Set of Columns: i.e. the default columns provided by journotwit or a single column containing all the tweets. Useful if you mess things up and you want to start over again.
Preset Column: choose from many different columns according to your needs, from different tweet types to memes (#followfriday, #musicmonday, etc.).
Custom Column: create your own personal column, according to your specific needs.

Because the overwhelming majority of my readers is composed by geeks, I’ll just describe how to create a custom column, so that you can fully understand the power of this tool, in the right hands.

Adding a custom column

The creation of a custom column doesn’t take long, but there are quite a few things you can configure. First off, you have to specify whether you want the column to collect local or global tweets: local means the people you follow, while global means everyone on the planet. Simple enough.

Then comes the juicy geeky part: search terms and tags. Simply type a valid Twitter Search query in the textbox, so something like this:

from:jonobacon OR #ubuntu -jaunty

…will hopefully fetch all tweets posted by @jonobacon or tweets about Ubuntu, but not related to the Jaunty Jackalope release. You can also add more text box and thus perform more search queries within the same column.

Then you can filter by tweet type, enabling or disabling Statuses, Visual, Links, Audio, ReTweets and Chatter. Useful to remove the noise (if you follow @brentspiner, make sure you disable ReTweets…).

Finally, you only have to configure a few more settings:

Whether you want to be alerted with a beep when there are new tweets in this column.
Whether you want the column to display tweets, a tag cloud or even an image slideshow.
The name of the column.

That’s all. Simple enough. As a side note, the “No-Mention” column is nothing but a custom column in disguise: if you try to edit it, you’ll see it’s nothing but a search for “username -@username -from:username”.

Other features and advanced settings

For the tweakers, JournoTwit also exposes the a set of global settings you can modify to enhance your experience or disable annoying behaviors (depends how you look at it):

Behavior

Unhide column when new tweets arrive? (default: yes)
Hide columns on Mark as Read? (default: yes)
Play alert sounds? (default: yes)
Animate when new items arrive? (default: yes)
Default #hashtags and search bar to a local search? (default: no)
Ignore Tag Coulds when marking all as read? (default: yes)
Ignore Slide Shows when marking all as read? (default: yes)
Warn when deleting columns? (default: yes)
Automatically translate tweets using Google Translate? (default: no)
Show “Did You Know?” messages on refresh? (default: yes)
Slide show transition time(s) (default: 5)

Display Adjustment

If you are unsatisfied by JournoTwit’s default look and feel, you can change the fond size, the color theme (there are 18 possible choices) and even match the color of the icons with the current theme.

Black Listing

Straight from the contextual help:

“Here you globally black list a #hashtag, such as #microsoft, or a search phrase such as “Windows 7”. Remember to separate them with a space and that you can block on a per column basis too."

This is just what you need when you want to filter out pointless tweets. Use with care though!

Conclusion

Maybe it’s just me being a geek, but I think JournoTwit nailed it when it comes to making Twitter more productive: everything just works, and fast, unlike some of its more feature-boasting competitors. I have been using it on a daily basis for weeks, and I’ve never missed a single tweet since (unless I explicitly wanted to do so).

That being said, there are a few small features I’d like to see:

I’d like to be able to mark single tweets, not entire columns, as read. In this way, when I go on vacation and come back, I can catch up with unread tweets more gradually, like I do with Google Reader.
I’d like to use shortcut keys to navigate the interface, like with Google Reader.
I’d like to configure tweets so that they only show up in one column, not in more than one (for example in Chatter, My Feed, and Mentions at the same time).
Support for Twitter Lists…

I already told @spode about some of these, and he said he’ll look into it, we’ll see what happens. Anyhow, just give it a try, and see if you like it!

A closer look at Komodo Edit

h3rald@h3rald.com — Sun, 25 Nov 2007 06:23:00 -0000

What’s your favorite code editor?

This is a common question which still keeps popping up on community boards, mailing lists, comments, etc. Every programmer who writes code in some programming language normally has an editor of choice. Being realistic, most of the times it’s not only one program but several, depending on the language and on the features needed at the time.
Yes, great, but how do you choose your favorite editor?
By trying a lot of them of course: that’s what I’ve been doing since I started programming. Lately I’ve become fond of Vim (or better, gVim), although I have a few other editors I may recommend, e.g.:

Recently, however, I came across my n^th^ “what’s your favorite editor” thread and someone mentioned Komodo Edit. I knew of it already, actually, but I never had a chance to give it a proper try.
ActiveState is well known mainly for two things:

Their effort in offering a lot of commercial, windows-based Perl libraries
Their commercial IDE, Komodo IDE, which I personally tried quite a long time ago

The problems of Komodo IDE are foundamentally two:

It costs money – this is not great in a world where Eclipse and Netbeans are free
It is an IDE – which normally means ‘heavy as hell’.

It’s a matter of taste and needs: some people feel more confortable with using an editor for certain tasks, while sometimes an IDE may be the best solution, despite its potential slowness and bulk.

Komodo Edit solves the above-mentioned problems because:

It’s free
It’s an editor

Actually I wouldn’t call it an editor, because it offers quite a lot of features which are normally absent in editors — it’s something in-between, which definitely deserves a try.

User interface

Komodo Edit’s interface is one of the most clear I’ve ever come across. It is organized in four main areas plus the top menu and toolbar (which only has the buttons you REALLY need, unlike Notepad++’s 31-button-bar).

The central area is for editing, the left pane is for the Project browser, the right one is for the Toolbox and the bottom pane is for command output. Luckily enough, a set of handy little buttons is provided in the toolbar to show and hide each pane. I normally only use the Project Browser and the main editing area, leaving the Toolbox and Command Output panes hidden, but that’s up to you really.

Komodo Edit allows symultaneous editing of multiple documents, which means that it has tabs like most of the best editors out there. Furthermore, it is possible to have Komodo re-open the files edited in the last editing session, saving you a lot of time. Granted, a lot of IDE do, but this is NOT an IDE, remember? It’s an editor, or at least it is advertised as such.

In reality it has most of the power of a full-blown IDE though normally remaining under the 60K memory usage (out of 2GB of total RAM) — which is not that bad, considering what Eclipse and NetBeans need.
Out of all the extra features offered by Komodo IDE, the only one I truly miss is the Code Browser. Other than that, it quickly became my editor/IDE of choice (when gVim isn’t looking, of course…)

Lets find out why…

Editing features

Komodo Edit comes with all the editing features of ever average editor, so “[…] Code commenting, auto-indent and outdent, block selection, incremental search, reflow paragraph, join lines, enter next character as raw literal, repeat next keystroke and clean line endings on “save”.", it uses Scintilla as main editing component, which makes it a close relative of Notepad++, Notepad2 and of course SciTE when it comes to understanding languages and syntax highlighting.

This is good, of course, as Scintilla is an excellent editing component, but – I have to say this – not as good as Vim when it comes to syntax highlighting (nothing is as good as Vim though, so I can’t really complain!).
As you can see from the screenshot I included earlier on, it is possible to change the default black-on-white color scheme to something more stimulating and energy-saving. This doesn’t mean I have to manually re-set the color of each element, it actually comes with a black-background scheme, which is handy.

Supported Languages & Syntax Highlighting

Komodo Edit supports quite a few programming languages, which means it can recognize the most common ones (C, C++, Java, PHP, Python, Ruby, Perl) but it also smart enough to notice framework-specific syntaxes like Smarty, Django or RHTML.
Like every scintilla-based editors, it sports a nice HTML multi-highlighter which allows you to keep putting all the Javascript, CSS and server-side scripting into the same file, GREAT! No, you probably wouldn’t want to do that, but it’s a good thing to have.

Auto-completion

What makes this editor a really handy tool is its auto-completion capabilities. Why? well, because it supports:

Ruby: require, class modules (on . and ::), class variables and methods, method calltips.
Python
Perl
PHP
XSLT
XML
Tcl

Why is this so special? Well, because if you want this you normally need an IDE, and if you want it for Ruby AND Python AND Perl your choices are very limited.
I played with the Ruby auto-completion features quite a bit, and I’m actually really impressed.

It can suggest what to write through calltips after a require directive, after a . and a ::, and hitting CTRL+SPACE (or just tab if you configure the right option) will auto-complete what you’re writing, be it a language keyword, a variable of any kind or a method. Additionally, it comes with “soft characters support”, which means it will try to match ( [ { when possible, allowing you to write over the completed character if you want to.

Syntax Checking, Vi/Emacs key bindings and code folding

The heading is almost self-explanatory here: Komodo Edit supports code folding (you can even chose the folding chracters!) for all its supported languages, syntax checking (it can detect and display errors before running the script, very useful in Ruby, and hard to find in other editors) and Vi/Emacs key bindings.
It also offers “vi emulation”, which is constantly improving through every release – so you won’t forget Vim’s keys when I don’t use it: very nice, and again, hard to find in other editors, nevermind IDEs!

Projects and Tools

The Project Browser and Toolbox are two features which are normally presents in IDEs and which are normally not found in editors. Granted, some editors like PSPad do indeed have “projects”, but most of the others don’t. A project in Komodo Edit is normally composed by:

Some Live Folders including all the source files of the application you’re developing
An optional set of tools, commands, snippets and other goodies which may help you coding that particular application.

It is possible to create these tools by clicking the Add Item to current project button in the Projects toolbar or by clicking the Add Item to Toolbox button just over the toolbox.
Either way, you can choose among the following:

File/Remote File

Add a file on your local machine or a remote file hosted on a server you configured via Edit > Preferences > Servers. Supported protocols are: FTP,FTPS, SFTP, SCP.

Folder

This will not link to a folder and import local files in Komodo, instead, it will create a container for other tools, not for local files.

Live Folder

On the contrary, a live folder links to an actual folder on your machines, and all files and directories inside it are automatically imported. Note that it is possible to choose to include or exclude certain files or directories from being imported by right-clicking the folder and editing its properties.

Command

This will create a shortcut to a command to execute. It is possible to specify its parameters, the directory where it will be run, whether its output will be displayed in the Command Output or in a newly-spawned console window, etc. etc.

Snippet

TextMate fans will love this. Komodo will let you create a code snippet for later use. Furthermore, it is possible to specify “tabstops” to automatically select certain words when tab is pressed (so you can effectively fill in only particular fields of the snippet).

Template

A simple but effective way to create code templates.

Url

A link which will open a given url.

Open… Shortcut

This will popup an Open File dialog in a direcxtory of your choice.

Macro

For those who require even more customization, it is also possible to record or code macros in Python or javascript, and then configure their key bindings and triggers.

Custom Menu/Custom toolbar

This was really impressive. Sure you can group all your tools in folders in your current project or you can create them in the toolbox to make them available all the time, but it would be cool if the editor itself could make those tools available by default in a toolbar or a menu. Well, that’s possible: all you need to do is create a new custom menu or toolbar and populate it with your tools by drag and drop. Very, very nice!

This mini-framework to create custom tools can be used to create project templates in a blink. To show this concept, Komodo Edit comes with a Rails template with almost all the tools you need when coding a Ruby on Rails application:

Create/delete database
Generators
Migrations
Run server
Example view snippets

Conclusion

Komodo Edit is an interesting editor which offers a lot of features which are normally only available in IDEs at a smaller memory footprint. Granted, when I want to jot something down I still prefer to fire up gVim, but for a serious coding session, Komodo is the right choice.
Its close attention to details and its ease of use make developing an even more pleasant experience, although perhaps there’s still room for improvement, if you’re really fussy.

Sometimes when you start getting used of its advanced features, the lack of a code browser seems a real shame, but after all, ActiveState must sell their IDE to someone at some point!

Another thing which I would have liked is the ability to create new syntax highlighting schemes, which Notepad++ and VIM both offer. Whydoes it bothers me so much? Well, because I’d like a Textile syntax highlighting scheme for writing blogs and articles. VIM and Intype seem to be the only ones which offer it out-of-the-box, but there’s no trace of it in Komodo Edit.

Actually it is possible to create new syntax highlighting schemes and extend Komodo Edit via XUL extensions, exactly like Firefox. After a closer look, the Django syntax was added in this way.

Unfortunately there aren’t that many extensions available for Komodo Edit, yet, but the Activestate Community seems very active, so you never know. Additionally, the recently-started OpenKomodo is a new project created by ActiveState to “[…] create an open source platform for building developer environments. ActiveState has open-sourced elements of Komodo Edit, a free multi-language editor for dynamic languages based on Komodo IDE, to create the Open Komodo code base”. A new competitor for Eclipse and Netbeans will be available soon?

RedBook - A simple Ruby program for your daily logging needs

h3rald@h3rald.com — Sat, 29 Sep 2007 12:05:00 -0000

Logging your daily activities is important. If you don’t believe me you’d better check at least these three posts on LifeHacker, which feature different scripts and applications:

I had a look at each one of them, and I believe they are quite useful, although I didn’t really find what I was looking for. Why? Well, for example:

I don’t believe a GUI is necessary — you’d better off with just a shortcut key or command to run from Launchy or QuickSilver, that’s much faster.
They just log timestamped messages on a file, there’s no real way to search through them and display them except by using a test editor
They are Windows only — not that it matters for me, but others may not be happy about it.

That’s why I thought I’d roll out my own: meet RedBook.
Let me say it’s nothing fancy: I’m not a full-time programmer but I do like playing with Ruby during my lunch breaks at work, so that’s why RedBook is just a humble, tiny Ruby script. This automatically makes it cross-platform: you can install Ruby very easily on Linux & alikes, Mac OS X and Windows. Furthermore, if you are on Windows and for some weird reason you don’t want to install Ruby, you can just try out the packed EXE file (made with RubyScript2Exe) — it’s about 2MB, but you won’t need anything else.

How It Works

The program uses two YAML files, one for configuration, which must reside in the same directory as redbook.rb (or redboo.exe) and one for the log itself, which you can place anywhere, provided that you edit the configuration file accordingly. For information on how to install RedBook and how to configure it, you can check the manual.html (powered by TiddlyWiki) file provided with the program or browse it online.

When started, RedBook will load both the configuration file (config.yml) and the whole log file into memory — it’s not a big deal, considering that they are only text files after all. I did a test with a log of quite a few MBs, and it was fine.

You can then start input commands right away, following a few sample rules. RedBook has a (very) rudimentary parser which is able to detect keywords, i.e. alphabetic strings prepended with a colon. RedBook commands look like this:

:log This message will be logger :tags tag1 tag2
:select :last 15 :since January
:save /home/h3rald/backup.yml

If everything goes OK, RedBook will reply with some sort of response, an acknowledgement, a list of messages, etc.

Simple.

Here’s basically what the program can do:

Log any message to the main log file. Messages can be tagged with one or more tags and will be automatically timestamped.
Load/display a list of logged messages on the screen. It is possible to filter messages by specifying a time span, a string to search in the message text, or a list of tags.
Dump loaded messages to a TXT, CSV or YAML file (you can even backup your log saving it to another YAML file in this way).
Calculate the time elapsed between two or more tasks. Time will be displayed in years, months, weeks, days, hours, minutes and/or seconds as necessary.

How? Here’s a short tutorial…

A Quick RedBook Tutorial

Let’s assume you are able to run RedBook on your system by now (if you can’t find some of the gems which are required for it, you can download them packed in a ZIP file from here).

Here’s what happens when you start the program:

-
RedBook v0.1 – Copyright © 2007, Fabio Cevasco
-
>> Loading config file…
>> Config file loaded.
>> Loading log file…
>> Log file loaded.
>> Ready.
RedBook >>

Good. Let’s start logging something then. Just use the :log keyword, followed by a message, and then you can also add the :tags keyword followed by space-separated tags, like this:

RedBook >> :log My first message :tags test
>> Logged.
RedBook >> :log This is another message
>> Logged.
RedBook >> :log This is another message :tags test another_test
>> Logged.

Try waiting a few seconds between each message. These three messages will be appended to the log file. You could open it in an editor, but it’s normally easier to display them directly inside RedBook, like this:

RedBook >> :select
1 Sat Sep 29 2007 – 09:09:32 PM My first message [test]
2 Sat Sep 29 2007 – 09:10:51 PM This is another message
3 Sat Sep 29 2007 – 09:11:45 PM This is another message [test][another_test]
>> 3 messages loaded.

Easy. What if you have hundreds of messages? Well, the :select operation can take an optional search string, or you can tell RedBook to load only those messages tagged with one or more specific tags, like this:

RedBook >> :select :tags test
1 Sat Sep 29 2007 – 09:09:32 PM My first message [test]
3 Sat Sep 29 2007 – 09:11:45 PM This is another message [test][another_test]
>> 2 messages loaded.

Or you can use the :from and/or :to keywords to specify a certain time frame, like this:

:select :tags test :from ten minutes ago

:select :from last week :to 2 days ago

RedBook includes a very nice “natural language date/time parser”, Chronic which is able to convert sentences like the following into Ruby Time objects:

6 in the morning
friday 1pm
sat 7 in the evening
today
yesterday at 4:00

It’s not perfect (and it’s in pre-alpha as well), but it does the job, for what I can see, and it makes it very easy and fast to specify timeframes.

After executing a :select command, two other operations can be performed on the loaded messages: :calc and :save.

:calc calculates the exact amount of time elapsed between two or more tasks. Do you remember the numbers on the far left of each message? Think them as temporary IDs for the actual messages, and you can use them to select specific tasks when executing the :calc operation:

RedBook >> :calc 1 3
1 Sat Sep 29 2007 – 09:09:32 PM My first message [test]
- 2 minutes and 13 seconds.
3 Sat Sep 29 2007 – 09:11:45 PM This is another message [test][another_test]

Similarly, if no IDs are specified, :calc calculates the time difference between each message and the previous:

RedBook >> :calc
1 Sat Sep 29 2007 – 09:09:32 PM My first message [test]

- 1 minute and 19 seconds.

2 Sat Sep 29 2007 – 09:10:51 PM This is another message

- 54 seconds.

3 Sat Sep 29 2007 – 09:11:45 PM This is another message [test][another_test]

Finally, you can save loaded messages to a TXT, YAML or CSV file, as follows:

RedBook >> :save log.txt
>> Saving…
>> Saved dataset to “log.txt”

h
If you want to backup your log, you can load all messages and then save them to a YAML file. Maybe in this case you want to append the :silent keyword to the :select command, so that messages won’t be displayed on the screen.

Conclusion

RedBook is just a simple program: it suits my needs for now, but of course there’s roo for improvement. If you have some useful suggestions, or you want to contribute in some way, feel free to contact me!

Home Page | Development | Download

Introducing RedBook (and the new Code section)

h3rald@h3rald.com — Sat, 29 Sep 2007 00:12:00 -0000

I’m somehow pleased to announce the opening of a new section on this site. Nothing too big actually, it’s just a page with a few (one for now) brief descriptions of open source programs and scripts I made and I’d like to share with my readers.

Don’t expect fancy stuff: (luckily) I don’t code for a living, I code for pleasure and I code small things. Lately I’ve been trying to write a small Ruby program able to log my daily activities and also display them in a pleasant enough way, so I started using my lunch breaks at work more constructively and I came up with RedBook an interactive command-line program written in Ruby.

Main features:

Log timestamped and tagged messages to a single YAML file
Load and display messages containing a certain string, or certain tags or within a time frame.
Calculate the time elapsed between two or more tasks.
Export loaded messages to YAML, TXT or CSV format.

All done via command line via simple commands:

:log This is a test message :tags test

:load :last 30 :from last week

:timecalc 2 5

:save test.txt

Thanks to the wonderful Chronic library, you can specify your time frames using natural language expressions like “8 in the morning”, “this tuesday”, “last month” and so on…

Here’s a screenshot showing RedBook in action:

RedBook is of course free, open source software licensed under the terms of the BSD license. It can be installed on any machine able to run Ruby and there’s also an EXE version for the lazy windows folks who don’t want to install Ruby, packed with RubyScript2Exe. A more in-depth article explaining how RedBook works ~~will hopefully be completed soon~~ is available :here

Tweaking Windows Explorer

h3rald@h3rald.com — Fri, 29 Jun 2007 09:28:00 -0000

If you asked me what file manager I used on Windows, up to a month ago I’d have answered something like: A43 or CubicExplorer, for sure anything but Windows Explorer.
Well, it turns out that I had to change my mind after all…

There’s a multitude of “Explorer Replacements” which aim to be more feature-rich, more user-friendly, less bloated than Bill’s favorite, and I indeed tried quite a few of them, mostly the free ones of course.

The only problem is that whenever I got close to choose “the one”, I noticed that there always was one or two features missing somewhere, which were present in another and vice versa. Additionally, to be totally honest, the level of integration with Windows and other applications was never complete.

These are a few stupid, silly things which tend to be lacking or at least are not 100% functioning — not in all the file managers I tried, but at least in some:

Environment variables integration — Alternative file managers normally are not able to parse Windows environment variables.
Icon overlays — Some file managers can’t render Subversion’s icon overlays.
Strange context menus — Sometimes right-clicking on a file or a directory may not open the standard Explorer context menu
Open Folder — Third-party application allowing you to open a folder, will always open it through Windows Explorer
auto complete address bar….
Special Folders — Some special folders (e.g. Control Panel) are now accessible using alternative file manager, but others may not (e.g. Network Connections). At any rate, typing “Control Panel” will not open the Control Panel in an alternative file manager.

What about trying to “patch” Windows Explorer instead of using another program altogether?

I never really thought of that until LifeHacker featured the QT TabBar shell extension by Quizo. That was my epiphany: I immediately thought I was going to “patch” Windows Explorer using a few really useful (and free, of course) Windows Shell Extensions.

Here’s my file manager now:

Yes, it is Windows Explorer, with just a few addons:

QT TabBar (Requires .NET framework 2.0)
QT Address Bar (Requires .NET framework 2.0)
QU ToolBar 2 (Requires .NET framework 2.0)
NTFS Link Shell Extension
FileMenu Tools

Explorer Tabs

For an instant gratification, download and install QT TabBar first, logoff and logon again for the changes to take effect (or install all the other extensions first, otherwise you’ll have to re-logon after installing nearly each one). As the name implies, this will add tabs to Windows Explorer, just the way you expect them to be: clean, stable and really smart.

After the installation, right-click anywhere in an existing window toolbar and enable the QT Tab Standard Buttons and voilá, a the tab bar will appear. You can click CTRL+N a few times to open new tabs. Take your time to position QT TabBar wherever is more convenient for you.

Whenever you’re ready, take your time to configure QT TabBar (right-click on it and choose “Options”) General options:

…and at least the Window options:

How do you create tabs? There are many different ways to do it, choose whatever is best for you:

CTRL+N keyboard shortcut
Drag and drop a folder or a shortcut on QT TabBar – Even from other applications!
Right click on an tab and choose “Clone this”

Other features? Sure, there’s plenty more, e.g.:

Single Instance — By enabling the appropriate option, you’ll only have a single instance of Windows explorer, everything will be opened in new tabs. A must.
Rearrange tabs — Tabs can be rearranged by dragging and dropping.
Lock/Unlock — Lock a tab (right-click > Lock) to avoid closing it accidentally.
Folder Memo — It is possible to set a “folder memo” for a specific folder, which can be visualized and edited afterwards (right-click > Memo for this folder).
Folder Password — Password-protect a particular tab (right-click > Set Password…).
Picture Preview — Preview image files simply by hovering on them with the mouse.
Groups — It is possible to group more tabs together by adding them to groups (right-click > Add to Group > [Group Name]) which can be configured via the QT TabBar option dialog.
Keyboard Shortcuts — Open the folder in which you unzipped QT TabBar and run QTShortcutKeyEditor.exe and enable as many as you like (they are too many to list hered), the possibilities are endless.

Customizing Toolbars

Windows Explorer allows users to choose the buttons which will be displayed in the Standar Buttons toolbar. To do so, proceed as follows:

If necessary, enable the Standard Buttons toolbar by selecting View > Toolbars > Standard Buttons from the Explorer menu.
Select View > Toolbars > Customize from the Explorer menu.
Choose the buttons you want to display.

I picked the following:

i.e.: Up , Refresh, Search, Folders, History, Favorites, Undo, Delete, Cut, Copy, Paste, Properties, View, Map Drive and Disconnect. It’s really up to you what you choose really.
I placed this toolbar right under the menu, on the top-left side.

Right next to it I placed some of the buttons available for QT TabBar:

Groups — Load an existing tab group (configurable in the options)
Recently Closed — open recently-closed tabs
Applications — run custom applications (configurable in the options)
Close — Close current tab
Lock — Lock current tab
Topmost — Force explorer to stay on top of other windows

Then I decided to enable the Links Explorer toolbar, which can be configured to display bookarks and shortcuts placed in the Favorites > Links folder. As shortcuts, I dragged each drive available on my system and voilà: poor man’s Drive Toolbar! Unfortunately, unlike in the proper drive toolbars offered by alternative file managers, all drive shortcuts will remain there (with a red question mark) even when the drive is not connected to the system. I can live with that.

Right to the address bar (we’re going to substitute it in the next section though), I decided to place QT ToolBar 2, which is available – hear, hear – after installing the QT ToolVar 2 extension.
At first it looks like a search filter toolbar, and yes, it can be used for this as well. Just type .jpg in the search box and it will show only the JPG files in the current folder. Easy enough.

Of course there’s (much) more to it. A search helper is provided (click the little arrow pointing downwards at the end of the toolbar and select Search Helper) to perform more complex searches:

Additionally, ToolBar allows you to display a handy copy file name/path button and up to two buttons to access two applications you use frequently. To configure them select Option from the dropdown menu accessible at the end of the toolbar, and configure your applications like this:

I choose the Command Prompt and the A43 file manager. The cool thing is that you can enable the “arguments for user application” and the the file path (if a file is selected) or the folder path will be passed automatically to the application, so my A43 file manage will open in the current folder. If you want to have more than two custom applications at your fingertips, all you have to do is to configure as many as you like in the Options tab of QT TabBar, and they’ll become available via the Applications button.

Two little utilities can also be used through QT TabBar 2, MD5, which instantly calculates the MD5 checksum of the selected file:

and Folder Analyze, which finds out the size of the current folder and how it is distributed across folders and files. A bit like a little WinDirStat but for the current folder only. It looks like this:

Navigation Improvements

Is there any way to make navigation through folder easier in Windows Explorer? Yep, more than one:

Breadcrumbs

Vista offers a Breadcrumb Bar, and XP doesn’t. Quizo fixed it of course, with his QT Address Bar, which brings breadcrumbs navigation to Windows XP. Use it as an Explorer address bar replacement. By default breadcrumbs are displayed:

…allowing you to navigate through your folder three within submenus without changing the current directory. Clicking it toggles the standard path:

A nice thing to have.

Take back your Favorites!

I don’t use IE, I use Firefox, Opera, even Safari sometimes, but not IE unless I’m forced to do so. Hence I hardly ever used Internet Explorer’s Favorites, an I almost forgot about it, until I decided to begin tweaking Windows Explorer, and I (re-)discovered that Favorites are shared between the two… What’s that got to do with anything? Well, you can simply put Favorites to good use and use them to store only Windows Explorer folders.

Simple and effective: click the Favorites button in the toolbar, they’ll be loaded in a sidebar on the left, then simply drag a folder to QT TabBar and you’ll get there.

Wormholes, anyone?

Finally, NTFS Link Shell Extension does something totally different altogether: it can be used to create hardlinks (a bit like Unix symlinks, but for NTFS drives only) junctions and symbolic links (Vista only). For an explanation of what each object is, refer to the explanations provided on the shell extension homepage. For our purposes, we’ll just use junctions now:

bq.“[…] Junctions are wormholes in the tree structure of a directed graph. By browsing a Junction a maybe far distant location in the file system is made available. Modifying, Creating, Renaming and Deleting files within a junction tree structure operates at the junction target, i.e. if you delete a file in a Junction it is deleted at the original location. […]”

Consider the following example.

1) Create a directory called “Gateway” in C:\.
2) With NTFS Link Shell Extension installed, right-click a directory “far, far away”, on any of your NTFS drives, for example D:\My\Very\Long\Path\MyDirectory, and select “Pick Link Source”.
3) Go back in your Gateway folder, right-click and select “Drop As > Junction”. A folder with a small chain overlay will be created.
4) You will now be able to access all the contents in D:\My\Very\Long\Path\MyDirectory directly from C:\Gateway.

Important – No, creating a shortcut is not the same thing. the path C:\Gateway\MyDirectory is an actual valid path, i.e. you can use it to attach files to emails, and going up one level in C:\Gateway\MyDirectory will take you simply to C:\Gateway\, not to D:\My\Very\Long\Path\.

Both at home and at work, I use a “Gateway” folder containing junctions leading to commonly-accessed directories, and this speeds up navigations a lot. Just remember to delete junctions “properly” (right-click > Delete Junction), not like an ordinary directory… ;-)

De-cluttering the Context Menus

Right now our Windows Explorer interface has been streamlined, folder navigation is easier, but there’s still room for improvement. Where? Well, in the contex menus of course.
I spent ages trying to figure out an easy way to remove unnecessary or unwanted entries from the menus which appears on a right-click. Yes, they can be removed by fiddling with the Windows Registry, but that’s not exactly user-friendly, is it? Now there’s an easy alternative: FileMenu Tools.

This nifty little utility allows you to remove rubbish from your context menus and add new entries as well. Entries are grouped by file type and can be enabled or disabled with a single click. Unfortunately I was not able to disabe some of them, probably due to restriction on my computer at work.

Once the rubbish is gone, perhaps you can even evaluate the possibility to add some more. I chose to enable just Attributes and Advanced Renamer, but there are many more predefined commands (each with its own pretty icon) you can choose from:

Synchronize Folders
Extended Delete
Find And Replace
Delete Locked File
Delete and no move to Recycle Bin
Change Icon
Run with Arguments
Command Line From Here
Split/Join File
Copy/Move to…
Copy Name/PAth/Content
Change Time
Register/Unregister DLL
Create new folder

Something missing? Well, you can always create your own entry, if you like!

Conclusion

I think that’s enough for you to give Windows Explorer another shot — or at least it was enough for me anyway. Be aware that there are and hopefully there will be more Windows shell extensions able to do neat things: 7-Zip, Notepad++ and Cream all add very useful context menus, and don’t forget TortoiseSVN, if you are a Windows-based developers (yes, they do exist).