SpringSource acquires hyperic. It is astonishing to see how well SpringSource played these last two years. While integrating various other products and frameworks SpringSource mastered the art of making a consistent Java alternative enterprise offering. I spent little time on monitoring and scaling, but I think it make sense.
SpringSource is kind of safest (specially considering the uncertain Oracle-SUN strategy), efficient way of doing lean good Java apps. I think its business model should be studied in depth. All their products are pure open source. Great documentation is provided (good affordable documentation is a key for community acceptance). So training, conference and expertise is their single financing at this stage. It is funny to see how SUN could never build a decent software service stream of revenues while alternate vendors did...
I hope they wille give it a shot in the RIA dance. Or do we have to consider that BlazeDS (Flex intergation) is their way to RIA ?
Tuesday, May 05, 2009
Monday, May 04, 2009
Wings - RIA - creation in Java or XML based language
I am quoting here the Wings framework web site.
"Popular frameworks like Sun JSF and Apache Struts rely on a heavy mix of different technologies like Java, JSP, XML & Tag libs. This implies that the association between the different media have to be maintained manually to be always in sync. Errors between these assoication implies bad suprises during runtime while pure Java-based frameworks like Wings can leverage of the compilers consistency checks."
I agree. Most XML based stuff are deferring checks at runtime. By the way this is my single regret with the Spring container and IOC. Fortunately the Spring container parses and validates the XML directive up-front when the application context is created so most errors can be easily checked at start-up.
"Popular frameworks like Sun JSF and Apache Struts rely on a heavy mix of different technologies like Java, JSP, XML & Tag libs. This implies that the association between the different media have to be maintained manually to be always in sync. Errors between these assoication implies bad suprises during runtime while pure Java-based frameworks like Wings can leverage of the compilers consistency checks."
I agree. Most XML based stuff are deferring checks at runtime. By the way this is my single regret with the Spring container and IOC. Fortunately the Spring container parses and validates the XML directive up-front when the application context is created so most errors can be easily checked at start-up.
Wings
One more GUI Ajax RIA framework for Java. http://wingsframework.org. One good point compared to ZK - real open source with a lgpl license.
It becomes more and more complicated to choose one :-) Seriously I think this storm of gui frameworks is a clear indication of an issue. Java lacks a clear path for building confortable interface (dummy HTML is not good for real application building).
It becomes more and more complicated to choose one :-) Seriously I think this storm of gui frameworks is a clear indication of an issue. Java lacks a clear path for building confortable interface (dummy HTML is not good for real application building).
Sunday, May 03, 2009
Apache Pivot - What is it ?
So I spent a bit of time this week-end with Apache Pivot (release 1.1).
It is a completely new GUI toolkit / framework built on top of Java2D. So it is not Swing, nor native browser controls.
This puts it in my opinion in the category of LRTW software (Lets Reinvent The Wheels). As mentioned in my first post, rendition is far from perfect.
The provided elements are good for layouts and containers but lack for me a good rich text editor - I think to get Java really on the desktop and shining for little businesses we need a programmable Java written office suite.
The interface can be built in Java or using an XML Based language: WTKX. May be this is more comfortable for some designers... Event handling plumbing seems to be in Java only - there is an empty scripting section in the tutorial so may be there are possibilities to declare event handler.
Pivot also includes utilities for data-binding (the stuff we miss in Java since years and you could find in VB or Delphi since day 0) and JSON (data exchange over HTTP), internationalization is also covered (I did not check it).
So my first opinion: this seems to be an open Java based Flex a like solution.
Unfortunately the deployment of the Java plugin is not comparable to Flash and the rendition if far from what you would get with native controls -e.g with Ajax or GWT.
Event being handled locally, so Pivot falls short compared to for example ZK / Echo and I think Canoo. In the demo I could not see an example with multiple windows / MDI like interface.
The tutorial is still incomplete. My statement : good open source = good documentation (potentially charge it but please write it if you want to gain quick acceptance).
What's more I could not find an indication of which problems Pivot is supposed to solve.
At this stage I cannot imagine to use it for anything. The project is currently 'incubated'. Except if the project can jump very quickly in term of quality and server based event handling, I don't think it will make long run.
It is a completely new GUI toolkit / framework built on top of Java2D. So it is not Swing, nor native browser controls.
This puts it in my opinion in the category of LRTW software (Lets Reinvent The Wheels). As mentioned in my first post, rendition is far from perfect.
The provided elements are good for layouts and containers but lack for me a good rich text editor - I think to get Java really on the desktop and shining for little businesses we need a programmable Java written office suite.
The interface can be built in Java or using an XML Based language: WTKX. May be this is more comfortable for some designers... Event handling plumbing seems to be in Java only - there is an empty scripting section in the tutorial so may be there are possibilities to declare event handler.
Pivot also includes utilities for data-binding (the stuff we miss in Java since years and you could find in VB or Delphi since day 0) and JSON (data exchange over HTTP), internationalization is also covered (I did not check it).
So my first opinion: this seems to be an open Java based Flex a like solution.
Unfortunately the deployment of the Java plugin is not comparable to Flash and the rendition if far from what you would get with native controls -e.g with Ajax or GWT.
Event being handled locally, so Pivot falls short compared to for example ZK / Echo and I think Canoo. In the demo I could not see an example with multiple windows / MDI like interface.
The tutorial is still incomplete. My statement : good open source = good documentation (potentially charge it but please write it if you want to gain quick acceptance).
What's more I could not find an indication of which problems Pivot is supposed to solve.
At this stage I cannot imagine to use it for anything. The project is currently 'incubated'. Except if the project can jump very quickly in term of quality and server based event handling, I don't think it will make long run.
Saturday, May 02, 2009
Apache Pivot
So I am having a look at Apache Pivot 1.1. I started the demo app in my browser (IE 7 - I know, I know...) just using the applet scroll bar and the entire interface went scratched.
I cannot elaborate more at this time. But Swing is a conceptually nice, complicated and frequently disappointing. Java applications suffer very often of poor renditions problems ?
In fact Pivot is not Swing based (it may looks like Swing for developpers) but is built on Java2D. But I am already sure that it exhibits similar behaviors...
I cannot elaborate more at this time. But Swing is a conceptually nice, complicated and frequently disappointing. Java applications suffer very often of poor renditions problems ?
In fact Pivot is not Swing based (it may looks like Swing for developpers) but is built on Java2D. But I am already sure that it exhibits similar behaviors...
Friday, May 01, 2009
ROO ? From SpringSource
One more stuff on my 'to evaluate list'. I must say that the Spring lightweight container is one of the most pleasant encounter I did with Java stuff these last two years.
My statement : right to the point. Good design, good documentation absolutely non obstructive, this makes programmers productive, designs elegant and solutions flexible...
This is the perfect example of how much innovation in the Java landscape is powered more by third parties than by the official JCP/JSR and owner of the language (SUN - now Oracle).
SpringSource also acquired the company behind Groovy and Grails. Nice language and probably easier to learn than Ruby On Rails (for a Java programmer) and with I think a much better (in fact perfect) integration with Java.
Well it seems that SpringSource continues its walk to ease programmer productivity with a project called 'ROO'. The name is not definitive it seems to be centered around Maven and provide command lines wizard to assemble application. Release is expected for July 2009. Wait and see.
My statement : right to the point. Good design, good documentation absolutely non obstructive, this makes programmers productive, designs elegant and solutions flexible...
This is the perfect example of how much innovation in the Java landscape is powered more by third parties than by the official JCP/JSR and owner of the language (SUN - now Oracle).
SpringSource also acquired the company behind Groovy and Grails. Nice language and probably easier to learn than Ruby On Rails (for a Java programmer) and with I think a much better (in fact perfect) integration with Java.
Well it seems that SpringSource continues its walk to ease programmer productivity with a project called 'ROO'. The name is not definitive it seems to be centered around Maven and provide command lines wizard to assemble application. Release is expected for July 2009. Wait and see.
Thursday, April 30, 2009
Adding Velocity tools to a velocity context
I could not find a basic example on how to add the Velocity tools library (in my case the EscapeTool) so I can use these helping routinged in my templates.
So here is mine:
org.apache.velocity.tools.generic.EscapeTool escTool = new org.apache.velocity.tools.generic.EscapeTool();
velocityContext.put("esc",escTool);
you can then call in your templates $esc.html($text_to_be_escaped)
I hope it helps
http://velocity.apache.org
So here is mine:
org.apache.velocity.tools.generic.EscapeTool escTool = new org.apache.velocity.tools.generic.EscapeTool();
velocityContext.put("esc",escTool);
you can then call in your templates $esc.html($text_to_be_escaped)
I hope it helps
http://velocity.apache.org
Tuesday, April 28, 2009
MySQL Limit - how to simulate with Microsoft SQL Server 2005
Much less strategic that previous post, I am quicly cooking a primitive web interface for which I need pagination.
In MySQL the limit added to the select helps verymuch retrieving a block of rows then the next one for the next page etc
What about MS Sqlserver ? You may start with TOP (takes the first rows) or set rowlimit.
But how to start at a given row ?
I found a few desperately incorrect proposals for this (but at least one pointed me to the ranking functions) , here is my best shot:
"with ordtab as (select p_users_id , ROW_NUMBER() OVER (order by p_users_id) as rownumber from ph_users) select * from ordtab where rownumber >= 10 and rownumber <= 20"
Will give you the equivalement of limit 10,20.
The solution is based on what is in the Microsoft manual, too bad we don't have time to read that much manuals, however it is still much less elegant than the MySQL attribute (which is right to the point)...
In MySQL the limit added to the select helps verymuch retrieving a block of rows then the next one for the next page etc
What about MS Sqlserver ? You may start with TOP (takes the first rows) or set rowlimit.
But how to start at a given row ?
I found a few desperately incorrect proposals for this (but at least one pointed me to the ranking functions) , here is my best shot:
"with ordtab as (select p_users_id , ROW_NUMBER() OVER (order by p_users_id) as rownumber from ph_users) select * from ordtab where rownumber >= 10 and rownumber <= 20"
Will give you the equivalement of limit 10,20.
The solution is based on what is in the Microsoft manual, too bad we don't have time to read that much manuals, however it is still much less elegant than the MySQL attribute (which is right to the point)...
Tuesday, April 21, 2009
Apache Pivot
A small Dzone article about Pivot. Pivot is an incubating project from Apache to make Swing based RIA application.
http://java.dzone.com/announcements/pivot-11
Also the link to Pivot (not that much doc, so better read the Dzone article before).
http://java.dzone.com/announcements/pivot-11
Also the link to Pivot (not that much doc, so better read the Dzone article before).
Monday, April 20, 2009
GlassFish
Continuing thinking on this Oracle-Sun deal, I don't think GlassFish has that much future anymore...
It is too early to make good prediction. Some analyst see here a thread to Microsoft (I don't think so) others think that Oracle will drop the hardware line, few think they will valorize the software line (I don't think so either).
Wait and see. In my readings, there is however a consensus... not good for open source...I agree.
It is too early to make good prediction. Some analyst see here a thread to Microsoft (I don't think so) others think that Oracle will drop the hardware line, few think they will valorize the software line (I don't think so either).
Wait and see. In my readings, there is however a consensus... not good for open source...I agree.
Orcale & Sun - applications to disk ?
Quoting the Sun site:
"The acquisition of Sun transforms the IT industry, combining best-in-class enterprise software and mission-critical computing systems. Oracle will be the only company that can engineer an integrated system – applications to disk – where all the pieces fit and work together so customers do not have to do it themselves. Our customers benefit as their systems integration costs go down while system performance, reliability and security go up."
Like it or not, clearly that's a vision. The One of a monolithic vendor...
"The acquisition of Sun transforms the IT industry, combining best-in-class enterprise software and mission-critical computing systems. Oracle will be the only company that can engineer an integrated system – applications to disk – where all the pieces fit and work together so customers do not have to do it themselves. Our customers benefit as their systems integration costs go down while system performance, reliability and security go up."
Like it or not, clearly that's a vision. The One of a monolithic vendor...
Oracle to acquire SUN
Sun and Oracle today announced a definitive agreement for Oracle to acquire Sun for $9.50 per share in cash. The Sun Board of Directors has unanimously approved the transaction. It is anticipated to close this summer.
Is it good or bad ? I never though to such a deal. So this is my humble reactions on very hot news.
Oracle will now become a kind of new IBM, concentrating hardware and software. Not sure that's indicating a sign toward openess...
Today, I received one mail from Sun : "Need more scalability for your MySQL application". With the key message being : we own it and we know better how to run it on our boxes ! Hm, if we extend this to the new portfolio, not sure that's so pleasant for all other hardware vendors proposing Oracle software...
What for Java? I don't think Oracle did very great contributions to the community, however I must admit that they are great at charging for anything (the main difference with Sun) . Will Java continue to be free ? Oracle is used to charge developer tools. Time to check the NetBeans licenses and ownership ?
And what about MySQL? Oracle is buying its single open source competitor. Isn't that a monopoly ? Will innovation continue on this DB Engine? It is funny to see Sun being acquired while they just finished building their application stack... oh bad is live when all you did is good but too late...
My last thinking. Sun managed Java poorly but on the other side this commercial weakness has been sustaining a relatively open ecosystem. Now with Oracle and IBM face to face, the major Java love triangle is over and the fight is almost guaranteed. I am betting on competition by so called innovation that will result in incompatibilities... Not good.
With Java and Lamp (MySQL) at stake, the winner might be... Microsoft !
Is it good or bad ? I never though to such a deal. So this is my humble reactions on very hot news.
Oracle will now become a kind of new IBM, concentrating hardware and software. Not sure that's indicating a sign toward openess...
Today, I received one mail from Sun : "Need more scalability for your MySQL application". With the key message being : we own it and we know better how to run it on our boxes ! Hm, if we extend this to the new portfolio, not sure that's so pleasant for all other hardware vendors proposing Oracle software...
What for Java? I don't think Oracle did very great contributions to the community, however I must admit that they are great at charging for anything (the main difference with Sun) . Will Java continue to be free ? Oracle is used to charge developer tools. Time to check the NetBeans licenses and ownership ?
And what about MySQL? Oracle is buying its single open source competitor. Isn't that a monopoly ? Will innovation continue on this DB Engine? It is funny to see Sun being acquired while they just finished building their application stack... oh bad is live when all you did is good but too late...
My last thinking. Sun managed Java poorly but on the other side this commercial weakness has been sustaining a relatively open ecosystem. Now with Oracle and IBM face to face, the major Java love triangle is over and the fight is almost guaranteed. I am betting on competition by so called innovation that will result in incompatibilities... Not good.
With Java and Lamp (MySQL) at stake, the winner might be... Microsoft !
Sunday, April 19, 2009
Download Sless 11 - Novell organisation at its best
I am preparing an upgrade of our main prod server to the new Suse Linux Enterprise Server 11. The latest, greatest, etc stuff from Novell Inc (pardon me,my excuse, I was a Suse users years ago).
After a (somehow slightly complicated) password recovery I could finally start downloading my ISO images.
Great Novell sent me an e-mail "Thank you for trying SUSE Linux Enterprise Server 10 from Novell."
Cool ,they forgot the mention concerning Netware 2.15 !
My point, e-business is not more complicated than business - but not less...
After a (somehow slightly complicated) password recovery I could finally start downloading my ISO images.
Great Novell sent me an e-mail "Thank you for trying SUSE Linux Enterprise Server 10 from Novell."
Cool ,they forgot the mention concerning Netware 2.15 !
My point, e-business is not more complicated than business - but not less...
Friday, April 17, 2009
About Java, market, technology and attitude
I read this morning a small article from Timothy M. O'Brien
http://broadcast.oreilly.com/2009/04/an-up-and-down-week-for-java.html
The article highlights the relation between Sun open/not so open behavior and its current market situation. Also, I agree entirely on the desperate JavaFX approach.
Would buying Sun be good to IBM ? If stock continues to drop, probably. Would it be good for Java: I think yes. IBM has been a fair player in the open-source community since several year and is good to listen to market and make money with services.
Yesterday, I saw a smart and interesting article "Meet Sun Software Engineering Manager Masood Mortazavi, Part 2: Java DB, Project Models, and More".
This is plenty of smart thinking, and I am sure Sun is full of these very talented people. Considering the situation, I will pardon the self-congratulating style of the article. However I was damned shocked by the following paragraph :
"SC: What are the biggest issues that companies have getting the most out of Java technology-based applications?
http://broadcast.oreilly.com/2009/04/an-up-and-down-week-for-java.html
The article highlights the relation between Sun open/not so open behavior and its current market situation. Also, I agree entirely on the desperate JavaFX approach.
Would buying Sun be good to IBM ? If stock continues to drop, probably. Would it be good for Java: I think yes. IBM has been a fair player in the open-source community since several year and is good to listen to market and make money with services.
Yesterday, I saw a smart and interesting article "Meet Sun Software Engineering Manager Masood Mortazavi, Part 2: Java DB, Project Models, and More".
This is plenty of smart thinking, and I am sure Sun is full of these very talented people. Considering the situation, I will pardon the self-congratulating style of the article. However I was damned shocked by the following paragraph :
"SC: What are the biggest issues that companies have getting the most out of Java technology-based applications?
Mortazavi: I would point to problems related to a mishmash of technologies, the lack of ability to select properly, and a relative increase in nonstandard, de-facto programming platforms."
Well. I cannot agree. First, what is the sense of choosing and "open" solution to stick to a single vendor ?
Second, Sun pushed developers to these so called 'mishmash' by regularly delivering technologies not matching business expectations. Remember the first releases without JDBC, crappy AWT, poor Swing performance, have you saw there was no reporting solutions built-in, still no clean integration with desktop under Windows... In fact, Java has been saved by these 'mismatch' like (to cite a few) Eclipse, Tomcat, Spring, Hibernate,JasperReports, Struts etc
We must understand that a lot of mishmashers just tried to fill the gaps... and they continue and continue. Why is Sun then endorsing Ruby , Groovy... Why are web applications more often built with PHP than Java...
I will one more quote this article "Don't forget that billions of dollars of revenue have been generated and millions of people have been employed because someone at Sun Microsystems invented Java,..."
That's correct, but I am sure we could have done much more money if Sun started to listen more to the market and the developers needs. The Java platform remains a very attractive but not so productive platform with a difficult learning stage (again the mentioned article is worth reading). Look at how much shrink wrapped software are written in Java... How much solutions for small business... well we missed a lot of business opportunities because of blind smart people...
Thursday, April 16, 2009
UltraLigt Client - Canoo
Still searching for an easy way to build these Java apps quicly and smoothly.
I will surely put this on in my list : http://www.canoo.com
This is a pure Java toolkit/framework. Client side runs as normal Java app (including web start) or as an applet. All the logic (event handling) is on the server.
An interesting compromise I think.
I will surely put this on in my list : http://www.canoo.com
This is a pure Java toolkit/framework. Client side runs as normal Java app (including web start) or as an applet. All the logic (event handling) is on the server.
An interesting compromise I think.
Wednesday, April 15, 2009
JavaFX ... Sun did it again
I am currently reviewing various architectures / frameworks etc for building (more – if possible) easily database backed applications.
Ideally we would like to leverage or Java knowledge (we did a lot with Java but little WEB/GUI stuff).
The idea of being able to deploy within a browser or outside is of course appealing.
So I had a look at JavaFX.
I quickly spotted the next link… http://blogs.sun.com/tor/entry/javafx_text_component . Waouw the next generation bablabla will not use text right now ! Cool ! “Forget about any business apps” for now could be on the web site banner. Know a business database apps without a single text edit?
So announcing so much, and providing so little, that reminds me how Java itself was launched (without proper GUI, no database API, even writing a file was an issue). Yes they did it again.
This is not only a technical exasperation, I cannot catch this strange blind marketing attitude !
Java is relatively standard for business software but the client side has to rely on a crappy mix of HTML/JavaScript to look a bit more responsive and stay easy to deploy?
Wouldn’t it be smart from SUN to capitalize on this market needs instead of trying to compete with Adobe Flash with an under capable product?
The latest Java runtime is much less deployed than Flash, the platform seems less capable and imply learning from scratch a new (said to be easy to learn) language.
Why would Flash developers move to this FX stuff ?
Why would Java developers move to this FX stuff ?
Well no clear replies and at this stage I give up on that one…
Ideally we would like to leverage or Java knowledge (we did a lot with Java but little WEB/GUI stuff).
The idea of being able to deploy within a browser or outside is of course appealing.
So I had a look at JavaFX.
I quickly spotted the next link… http://blogs.sun.com/tor/entry/javafx_text_component . Waouw the next generation bablabla will not use text right now ! Cool ! “Forget about any business apps” for now could be on the web site banner. Know a business database apps without a single text edit?
So announcing so much, and providing so little, that reminds me how Java itself was launched (without proper GUI, no database API, even writing a file was an issue). Yes they did it again.
This is not only a technical exasperation, I cannot catch this strange blind marketing attitude !
Java is relatively standard for business software but the client side has to rely on a crappy mix of HTML/JavaScript to look a bit more responsive and stay easy to deploy?
Wouldn’t it be smart from SUN to capitalize on this market needs instead of trying to compete with Adobe Flash with an under capable product?
The latest Java runtime is much less deployed than Flash, the platform seems less capable and imply learning from scratch a new (said to be easy to learn) language.
Why would Flash developers move to this FX stuff ?
Why would Java developers move to this FX stuff ?
Well no clear replies and at this stage I give up on that one…
Sunday, May 04, 2008
Zoho
I discovered recently Zoho (www.zoho.com), a desktop application service provider. The tools are neat and expanding. You can do the classical office jobs such as word processing, spread-sheet and presentation, a mail app is still in beta and I haven't tried their database application yet.
There are also business apps coming: CRM and invoicing. The business model is not advertising or marketing based (I applause), basic usage is free while heavy / professional usage may be charged.
Globally it is a nice set of tools. In my opinion the tag-based filling system is interesting, yet it is difficult to see all type of document related to let say a project or a customer. Also I could not find an easy way to include a spread-sheet in a text - except may be like an HTML iframe. I admit, I am a novice user of their system.
A much more strategic issue is concerning technical integration and freedom. We miss a global model and a good bunch of standards to rent applications from multiple vendors and mix them to achieve an IT system.
Having said that, the worth part is in the terms and conditions. Well, read and think about it... Prices may change, free things may become costy and no warranty is made about your data, availability of services etc.
This is in my opinion, a key problem of this kind of business (the following it is not specifically related to Zoho). We have here an emerging industry rejecting most forms of liability, warranty and responsibility!
Obviously a remote ASP cannot guarantee that my ISP will do its job, but aren't they responsible if they garbage my data? Or may be should they think about charging for such extra-liability. Obviously the bottom line is made by insurance companies, only these can help creating professional ASP a reality. The model should be similar to life insurance and air transportation: a basic fee built in your ticket and more can be purchased if the customer wants to do so.
As I wait for that, I would recommend these services to implement at least a clean backup system (some ISP propose this for databases), so a user can retrieve all its data in a somehow common format.
There are also business apps coming: CRM and invoicing. The business model is not advertising or marketing based (I applause), basic usage is free while heavy / professional usage may be charged.
Globally it is a nice set of tools. In my opinion the tag-based filling system is interesting, yet it is difficult to see all type of document related to let say a project or a customer. Also I could not find an easy way to include a spread-sheet in a text - except may be like an HTML iframe. I admit, I am a novice user of their system.
A much more strategic issue is concerning technical integration and freedom. We miss a global model and a good bunch of standards to rent applications from multiple vendors and mix them to achieve an IT system.
Having said that, the worth part is in the terms and conditions. Well, read and think about it... Prices may change, free things may become costy and no warranty is made about your data, availability of services etc.
This is in my opinion, a key problem of this kind of business (the following it is not specifically related to Zoho). We have here an emerging industry rejecting most forms of liability, warranty and responsibility!
Obviously a remote ASP cannot guarantee that my ISP will do its job, but aren't they responsible if they garbage my data? Or may be should they think about charging for such extra-liability. Obviously the bottom line is made by insurance companies, only these can help creating professional ASP a reality. The model should be similar to life insurance and air transportation: a basic fee built in your ticket and more can be purchased if the customer wants to do so.
As I wait for that, I would recommend these services to implement at least a clean backup system (some ISP propose this for databases), so a user can retrieve all its data in a somehow common format.
Thursday, April 24, 2008
Office 2007
Office 2007 is a major interface change. Changing an interface that people are used to is always an issue. I admit, I hate it, its only purpose seems to fill more space on screens that are getting bigger...
Today I wanted to change the outline numbering of my paragraps, so paragraph numbering would look 1.1, 1.2, 1.3 - 2, 2.1 etc.
Where is that? First difficulty is finding the Modify Style dialog. The good old menus are gone...
Finally a right click on the Style Toolbox/TextBox on the ribbon (I did not first, this was calle the style gallery...) starts the dialog. Having spent one hour on this dialog (productivity gain announced by MS seems a joke), I could not changed the outline numbering.
Finally I Googled, and found http://kitss.wordpress.com/2007/08/08/word-and-numbering-2-outline-numbering/
.
Well, select your style on the Toolbox then click on the Drop-dow (multi-level level) in the group Paragraph and yes... there you can change this outline numbering!
Do you know other way of doing it? Let me know.
Today I wanted to change the outline numbering of my paragraps, so paragraph numbering would look 1.1, 1.2, 1.3 - 2, 2.1 etc.
Where is that? First difficulty is finding the Modify Style dialog. The good old menus are gone...
Finally a right click on the Style Toolbox/TextBox on the ribbon (I did not first, this was calle the style gallery...) starts the dialog. Having spent one hour on this dialog (productivity gain announced by MS seems a joke), I could not changed the outline numbering.
Finally I Googled, and found http://kitss.wordpress.com/2007/08/08/word-and-numbering-2-outline-numbering/
.
Well, select your style on the Toolbox then click on the Drop-dow (multi-level level) in the group Paragraph and yes... there you can change this outline numbering!
Do you know other way of doing it? Let me know.
Monday, April 14, 2008
Netbeans 6.x as a PHP IDE
As I am currently working with a PHP written CMS - Joomla - I tried the new beta plugin for Netbeans. With this extension you can create PHP projects and get code completion. Debugging in the IDE is possible provided you install the xdebug extension on the server.
As usual the Linux distribution is done through source code, it tooks me around 20 minutes to get the build process up and running. Debugging worked fine for simple application but for Joomla the debugger started but the resulting page was always blank.
I finally give up and used Wamp server on windows, there is worked perfectly. I still don't know why I can't get it up and running on my Linux machine. Among the difference : wamp comes with php 5 per default.
Links
The extension for PHP editing in Netbeans : php.netbeans.org.
Debugger for PHP : www.xdebug.org.
Apache + PHP + MySQL : www.wampserver.com.
As usual the Linux distribution is done through source code, it tooks me around 20 minutes to get the build process up and running. Debugging worked fine for simple application but for Joomla the debugger started but the resulting page was always blank.
I finally give up and used Wamp server on windows, there is worked perfectly. I still don't know why I can't get it up and running on my Linux machine. Among the difference : wamp comes with php 5 per default.
Links
The extension for PHP editing in Netbeans : php.netbeans.org.
Debugger for PHP : www.xdebug.org.
Apache + PHP + MySQL : www.wampserver.com.
Friday, April 04, 2008
Mail server crash
This week-end, our mail server crashed. I must admit this is only a half surprise, as strange noises were heard from this server since a few weeks. Unfortunately the largest (and newest – but still old) hard disk crashed, bringing our Exchange databases to the world of lost data.
In fact I have plans to replace this Microsoft Exchange 5.5 on NT 4.0 by something else since a few months. Since two years I am trying to consolidate our servers on a bigger Fujitsu-Siemens Tx300 S2 with Novell SLES 10.x and VMWARE server. So my idea was to find a free (or cheap) alternative to MS Exchange working on Linux.
There are many such applications on the market, all claiming to be the perfect replacement for the market leader (or at least the reference).
Open-Xchange
A month ago, I evaluated Open-Xchange for that purpose.
My first remark is about Open Source and business model. How can a company make money with free product ? There is a trend to provide a community version (with source code) and no support and to sell or rent an enterprise class product. What is usually unclear is which version of the Open-Souce code correspond to a stable product (if any).
In this case, you have the choice of a community product, an Express product and an enterprise product. Licensing of the open source, community version could be subject of an article as the Web GUI has not the same license as the rest of the code (smell bad). There is also a huge difference of price between Express edition (starting 5 users) and Enterprise (25 users). Also the pricing is under a yearly subscription basis (MS is priced for life) so a comparison need to be done on several years and take account how often you will upgrade (if you know..).
Finally I give up with Open-Xchange :
A customer pointed me to Scalix a few weeks ago. This is again an Open-Source Linux based challenger for Microsoft Exchange. There is also here some questioning about the Open-source licensing (you can find articles on the web - the Scalix licence could be not OSI approved - I think this requires a serious study), but at least you have a free community edition with all the functionalities – including Outlook integration (but for 10 users – enough for me).
I must admit the condition of test are not ideal, I created a VMWARE virtual machine with around 512 MB of RAM and OpenSuse 10.2 (I known 10.3 is out but 10.2 is in the Scalix supported list). It worked ! I have been first very impressed by the Web interface, it is very similar to a desktop application, you can even use drag and drop ! The installation of the Outlook connector went also fine, the functionality is comparable to the original MS Exchange. So I used this Outlook connection to load my personal contacts (around 400 records). Starting there, there has been a permanent disk activity and the web interface which was good but slow became very very very slow.
Again I admit the virtual hardware I provided is minimal. After a while, popup errors started on the GUI, at that stage the thing went unusable. I tried to extend the VM memory to 700 MB (more would result in swapping), I still noticed permanent disk activity on the server – related to the VM running Scalix.
Also that day I also tried to use my backup of the Exchange database, I could recover the directory, private and public databases but the MTA could not start – I did not backup that part !
After my Scalix investigation, I went to bed.
The day after, I restarted the Scalix VM. The thing was still very slow for the Web GUI but again more or less workable and the Outlook connector behaved very properly. So Scalix was my temporary saver.
Exchange is back
Surprise ! My colleague congratulated me for putting back the Exchange service ! Well, it appears that the MTA restarted by itself – there is a service agent permanently trying to restart other services. How it did, mystery ! Since day 0, I have always been impressed by the reliability of MS Exchange – at least the old version. Here again “bravo”.
Still digging in Scalix
The “by chance” recovered Exchange is not going to stay as our mail server. It is 10 years old and I am sure one day or another the other disks will fail. So I tried putting the Scalix software directly on SLES 10 instead of being hosted in a VM. Unfortunately my TX300 is a 64 bits machine, but the Scalix install process asked for some 32 bits libraries. I found an indication on a forum to install the 32 bits version of the library – it seems to be a know Scalix issue-, but I did not tried. Why ? Because with my mail server crashed I did not want to endanger our main remaining machine.
So I decided to buy a new entry level server, for mail purpose. My idea was to check first for Dell, I found the server page, small business etc and there I saw an arrow… “Start configuring” but I could not start anything (nor with Firefox nor with IE). So I give up. My opinion, one more damned Javascript stuff not working... Finaly I ordered a new Fujistsu-Siemens Primery Econel 100 S2. It is there, but not yet installed.
Will I keep Scalix? I don’t know, I am going to give it one more try with the new server. One thing , we remarked incompatibility between the web GUI and some version of IE (including version 7). Of course, I can install Firefox everywhere, but I cannot be sure to find it everywhere when I am outside of the office…
My conclusions for today
In fact I have plans to replace this Microsoft Exchange 5.5 on NT 4.0 by something else since a few months. Since two years I am trying to consolidate our servers on a bigger Fujitsu-Siemens Tx300 S2 with Novell SLES 10.x and VMWARE server. So my idea was to find a free (or cheap) alternative to MS Exchange working on Linux.
There are many such applications on the market, all claiming to be the perfect replacement for the market leader (or at least the reference).
Open-Xchange
A month ago, I evaluated Open-Xchange for that purpose.
My first remark is about Open Source and business model. How can a company make money with free product ? There is a trend to provide a community version (with source code) and no support and to sell or rent an enterprise class product. What is usually unclear is which version of the Open-Souce code correspond to a stable product (if any).
In this case, you have the choice of a community product, an Express product and an enterprise product. Licensing of the open source, community version could be subject of an article as the Web GUI has not the same license as the rest of the code (smell bad). There is also a huge difference of price between Express edition (starting 5 users) and Enterprise (25 users). Also the pricing is under a yearly subscription basis (MS is priced for life) so a comparison need to be done on several years and take account how often you will upgrade (if you know..).
Finally I give up with Open-Xchange :
- first, I found the handling of international locale very poor (I could not find an easy way to display date in European format yet having the interface in English).
- second, the support for Outlook is very basic with calendar sharing but e-mail is available as an IMAP/POP mail server instead of a unified/synchronized folder within Outlook (as Microsoft Exchange does).
A customer pointed me to Scalix a few weeks ago. This is again an Open-Source Linux based challenger for Microsoft Exchange. There is also here some questioning about the Open-source licensing (you can find articles on the web - the Scalix licence could be not OSI approved - I think this requires a serious study), but at least you have a free community edition with all the functionalities – including Outlook integration (but for 10 users – enough for me).
I must admit the condition of test are not ideal, I created a VMWARE virtual machine with around 512 MB of RAM and OpenSuse 10.2 (I known 10.3 is out but 10.2 is in the Scalix supported list). It worked ! I have been first very impressed by the Web interface, it is very similar to a desktop application, you can even use drag and drop ! The installation of the Outlook connector went also fine, the functionality is comparable to the original MS Exchange. So I used this Outlook connection to load my personal contacts (around 400 records). Starting there, there has been a permanent disk activity and the web interface which was good but slow became very very very slow.
Again I admit the virtual hardware I provided is minimal. After a while, popup errors started on the GUI, at that stage the thing went unusable. I tried to extend the VM memory to 700 MB (more would result in swapping), I still noticed permanent disk activity on the server – related to the VM running Scalix.
Also that day I also tried to use my backup of the Exchange database, I could recover the directory, private and public databases but the MTA could not start – I did not backup that part !
After my Scalix investigation, I went to bed.
The day after, I restarted the Scalix VM. The thing was still very slow for the Web GUI but again more or less workable and the Outlook connector behaved very properly. So Scalix was my temporary saver.
Exchange is back
Surprise ! My colleague congratulated me for putting back the Exchange service ! Well, it appears that the MTA restarted by itself – there is a service agent permanently trying to restart other services. How it did, mystery ! Since day 0, I have always been impressed by the reliability of MS Exchange – at least the old version. Here again “bravo”.
Still digging in Scalix
The “by chance” recovered Exchange is not going to stay as our mail server. It is 10 years old and I am sure one day or another the other disks will fail. So I tried putting the Scalix software directly on SLES 10 instead of being hosted in a VM. Unfortunately my TX300 is a 64 bits machine, but the Scalix install process asked for some 32 bits libraries. I found an indication on a forum to install the 32 bits version of the library – it seems to be a know Scalix issue-, but I did not tried. Why ? Because with my mail server crashed I did not want to endanger our main remaining machine.
So I decided to buy a new entry level server, for mail purpose. My idea was to check first for Dell, I found the server page, small business etc and there I saw an arrow… “Start configuring” but I could not start anything (nor with Firefox nor with IE). So I give up. My opinion, one more damned Javascript stuff not working... Finaly I ordered a new Fujistsu-Siemens Primery Econel 100 S2. It is there, but not yet installed.
Will I keep Scalix? I don’t know, I am going to give it one more try with the new server. One thing , we remarked incompatibility between the web GUI and some version of IE (including version 7). Of course, I can install Firefox everywhere, but I cannot be sure to find it everywhere when I am outside of the office…
My conclusions for today
- a lot of companies are saying they are Open Source, but they seem to have priced products and they are not always clear about it, it is still worth checking but you must think before going ahead.
- AJAX web based client applications are very trendy but I am still waiting for one that is nice and stable in production use.
- don’t forget to make good, regular, tested backups.
Subscribe to:
Posts (Atom)