Sunday 5 October 2014
Ali moves blog to a dedicated site
All future technology blogs will be posted at the following web address: https://www.alirobertson.net . Thank you for supporting this blog over the years. I will now be running an ICT education/research blog, providing free resources (e.g. survey data, papers, educational resources & advice) to those that need it.
Friday 30 March 2012
Mark Shuttleworth, Ubuntu creator, motivates students at the Isle of Man College
It was a whistle stop tour of his life from the time he sold his first business to Verisign, to making reality his ambition to become a space tourist! Finally, he spoke about how he created and manages the Ubuntu project. This guy is a real motivator and we have been fortunate to have had the presentation recorded for future training needs.
For the open-source community there is also a lot to learn regarding his vision for the future of software developments, particularly his current project that will provide 'easy to deploy', open-source cloud solutions. In an earlier blog, I wrote about private versus public Clouds and the likely requirement for more private versions as security risks mount. I'm not certain how the new software will fit into the private/public cloud debate, but I look forward to finding out shortly.
Many thanks to Mark for his motivational talk to our Manx students on the BA Computer Science programme and to those students that'll be taking the programme in following years. The presentation recording will be used for training and facilitating learning among our students for years to come; after all, it's good to be motivated by someone who has done it their way, and for the right reasons (that is, open-source creativity). Mark's passion for the open-source movement was strongly evident in his presentation, hopefully, it'll be infectious among our learners also.
Could you be our next motivational speaker or know of someone who could be? Please contact myself (alastair.robertson@iomcollege.ac.im) or Roger Cowin (roger.cowin@iomcollege.ac.im). The college provides various degrees (BSc Health and Social Care, BSc Computer Science, BA Business Studies and the MBA) whose students would always benefit from seeing practitioners speak about their experiences.
A special thanks to Manx TV (Paul Moulton) for recording the event. The follow-up interviews with Mark can be seen here and here.
Hey you, get off of my cloud…c:\hiding geek-speak from the users\☺
In the age of broadband connectivity new terms have emerged that help support new and ‘evolutionary’ technologies to become common place. Commonly asked questions, early on in the technology adoption cycle, usually focus around clarification of the new terms and services. For the techno-geek fraternity out there, this becomes opportunity to show off their knowledge that may even quicken adoption of the technology that presents itself. Sometimes terminology could never make real sense to novice IT users with even near experts struggling to understand new ways of working; but occasionally a new way of explaining complexity emerges that really does simplify matters. As windows became the watchword for an operating system, the term Cloud should become the watchword for doing ‘my stuff’ online.
I have been asked on numerous occasions to explain how the ‘cloud’ can influence the way we work; and also why it is called Cloud, and not, Space or Sky. The timeline to the development of cloud terminology is unclear but I suspect it was generated to help inform non-IT experts of a space where your ‘stuff’ and ‘stuff you do’ is stored or accessed by you but you don’t need to know where or how it’s done. Traditionally, people needing advanced IT services also required a firm understanding of the 4 layer model of networks, resources, middleware and applications. Falling asleep already? Then a better way to explain this is to describe the cloud as a space that is managed by IT experts who take care of all the difficult technological bits (e.g. updating software, keeping the network up, fixing failed servers); and you just get on and do your work, just like the halcyon dream of computing used to be but failed to deliver on. Just like the average motorist doesn’t need to know how an engine works, cloud users do not need to know how networks and data farms do what they do. Calling it something like cloud paints a marketable image in the mind of the average person that detracts thinking away from wires and computer boxes and ghastly terminology like Platform as a Service (PaaS) or Software Orientated Architecture (SOA). I bet the computer industry hopes that it may even attract them into adopting cloud services more readily. Crucially, they should just seamlessly interact with ‘information space’ without ever needing to understand the complex infrastructure and processes that provide their ability to do so. Just like windows moved us away from direct communication with hard drives by giving us pretty icons to click on instead of difficult code (e.g. C:\god this is complicated\inhell\etc)…cloud helps hides the real complexity underlying its structure; but this comes at a price, right?
Although the transition into graphically pleasing operating systems helped increase the adoption of PCs into homes and businesses around the world, simplification of online software service provision, via the cloud, may do something more sinister for those early adopters that have not read the small print carefully. Security is a major source of concern in cloud services and is an area impossible to escape from if you are considering using online services. Security is also semantically technical which, at least, will keep the techno-geeks like me happy☺. Lots of tricky terms emerge like cyber security, data centres, service uptime, hackers and firewalls; and there isn’t an easy way of escaping this kind of terminology without generalizing away from the real risks. Work coming out of the Science and Technology Facilitation Council (STFC) Daresbury, under the direction of Dr Rob Allan, suggests that if you do not want your information/data to fall into the wrong hands then you probably shouldn’t be using the cloud at all. Although this comment is in stark contrast to cloud industry views, risks exist in the cloud that go beyond YOUR abilities to control. After all, most people do not use sufficiently complex passwords that would otherwise protect them from computerized hack attacks; and a good deal of criminal activity occurs within companies by its employees meaning that your account could be hacked very easily from within the cloud supplier. Few averagely skilled computer users are aware that systems administrators within cloud suppliers have full access to your accounts and therefore data…in the ‘old school’ IT model where all data was stored and protected by your IT department, safety protocols could be embedded to factor in the risks posed by company employees. This is much harder to police in cloud environments where several supplier businesses may be linked together to provide a complete cloud service. For instance, cloud technology businesses may rely on server farms to host their apps that are then sold to you. This means your data could be accessible to multiple businesses without you even knowing. OK, there are approaches to encrypting your data at various levels, but the truth remains, managing your own data on your own servers is the most secure approach.
Beyond the risk, fortunately in the UK at least, we are governed by clearly stated data protection regulations. The Data Protection Act protects users by forcing cloud suppliers and cloud service purchasers, that are based in the UK, to become data controllers; but many cloud suppliers are located away from this protected area (e.g. US). Although all cloud companies will do their best to protect your data no one can predict when or if the cloud supplier will be sold. It can be sold, and yes, your data could be moved to a geographic location that is not data protected. If you are a UK based organisation using a cloud supplier whose data is moved in this way, if you store data about customers that is data protected (e.g. names and telephone numbers, sales invoices etc.); you’ve broken the law as you are likely to be the designated data controller. So how can we get around this tricky legal issue?
The Rolling Stones song Hey you, get off of my cloud was rather prophetic to the issue. Without throwing away the benefits that the cloud can bring is the development of the private cloud concept; that is, run smaller scale, non-global clouds tailored to local conditions. For example, a private cloud guaranteeing UK data protection at all times as part of the service level agreement to customers. Some larger organisations already run their own private cloud services and technology is now available that allows them to do this will relative ease (i.e. virtualization). This provides them most of the benefits of cloud with less risk. The offset however, is that you still need to manage your own servers which is costly. Remember that by owning the servers you are in full control of the data. For smaller organisations that don’t have the ability to host their own services the image is less clear. It could be that ‘localized’ application service providers (ASPs) will work within guaranteed data jurisdictions meaning that data is always legally protected. ASPs could form from internet service providers that would hold the key to success. They understand the context of UK Data Protection regulations already and so will have a handle on how to manage your data via software tools they will rent to you. It is unclear if or when such services will be developed and deployed however. It will also require that re-sellable software becomes available and at this time few exist outside of a few collaboration (e.g. Microsoft SharePoint) and customer relationship (e.g. Sugar CRM) packages. To be honest, cloud technologies are presently few and far between and the long-term viability of the cloud concept will only be sustained if more services become available over the coming years. We will also have to watch out for evolutions of the cloud. For instance, Google’s Chrome operating system is technically a hybrid of the cloud vision by allowing users to directly integrate with Google cloud. Given Google’s successful development and deployment of the Android handset technology this may be a space to watch out for. In mean time, we’ll just have to wait to see how the sky looks when the sun eventually rises. Only then will we be able to see how many clouds are on the horizon, global or localized. Or indeed, who will be prepared to inhabit them…
I have been asked on numerous occasions to explain how the ‘cloud’ can influence the way we work; and also why it is called Cloud, and not, Space or Sky. The timeline to the development of cloud terminology is unclear but I suspect it was generated to help inform non-IT experts of a space where your ‘stuff’ and ‘stuff you do’ is stored or accessed by you but you don’t need to know where or how it’s done. Traditionally, people needing advanced IT services also required a firm understanding of the 4 layer model of networks, resources, middleware and applications. Falling asleep already? Then a better way to explain this is to describe the cloud as a space that is managed by IT experts who take care of all the difficult technological bits (e.g. updating software, keeping the network up, fixing failed servers); and you just get on and do your work, just like the halcyon dream of computing used to be but failed to deliver on. Just like the average motorist doesn’t need to know how an engine works, cloud users do not need to know how networks and data farms do what they do. Calling it something like cloud paints a marketable image in the mind of the average person that detracts thinking away from wires and computer boxes and ghastly terminology like Platform as a Service (PaaS) or Software Orientated Architecture (SOA). I bet the computer industry hopes that it may even attract them into adopting cloud services more readily. Crucially, they should just seamlessly interact with ‘information space’ without ever needing to understand the complex infrastructure and processes that provide their ability to do so. Just like windows moved us away from direct communication with hard drives by giving us pretty icons to click on instead of difficult code (e.g. C:\god this is complicated\inhell\etc)…cloud helps hides the real complexity underlying its structure; but this comes at a price, right?
Although the transition into graphically pleasing operating systems helped increase the adoption of PCs into homes and businesses around the world, simplification of online software service provision, via the cloud, may do something more sinister for those early adopters that have not read the small print carefully. Security is a major source of concern in cloud services and is an area impossible to escape from if you are considering using online services. Security is also semantically technical which, at least, will keep the techno-geeks like me happy☺. Lots of tricky terms emerge like cyber security, data centres, service uptime, hackers and firewalls; and there isn’t an easy way of escaping this kind of terminology without generalizing away from the real risks. Work coming out of the Science and Technology Facilitation Council (STFC) Daresbury, under the direction of Dr Rob Allan, suggests that if you do not want your information/data to fall into the wrong hands then you probably shouldn’t be using the cloud at all. Although this comment is in stark contrast to cloud industry views, risks exist in the cloud that go beyond YOUR abilities to control. After all, most people do not use sufficiently complex passwords that would otherwise protect them from computerized hack attacks; and a good deal of criminal activity occurs within companies by its employees meaning that your account could be hacked very easily from within the cloud supplier. Few averagely skilled computer users are aware that systems administrators within cloud suppliers have full access to your accounts and therefore data…in the ‘old school’ IT model where all data was stored and protected by your IT department, safety protocols could be embedded to factor in the risks posed by company employees. This is much harder to police in cloud environments where several supplier businesses may be linked together to provide a complete cloud service. For instance, cloud technology businesses may rely on server farms to host their apps that are then sold to you. This means your data could be accessible to multiple businesses without you even knowing. OK, there are approaches to encrypting your data at various levels, but the truth remains, managing your own data on your own servers is the most secure approach.
Beyond the risk, fortunately in the UK at least, we are governed by clearly stated data protection regulations. The Data Protection Act protects users by forcing cloud suppliers and cloud service purchasers, that are based in the UK, to become data controllers; but many cloud suppliers are located away from this protected area (e.g. US). Although all cloud companies will do their best to protect your data no one can predict when or if the cloud supplier will be sold. It can be sold, and yes, your data could be moved to a geographic location that is not data protected. If you are a UK based organisation using a cloud supplier whose data is moved in this way, if you store data about customers that is data protected (e.g. names and telephone numbers, sales invoices etc.); you’ve broken the law as you are likely to be the designated data controller. So how can we get around this tricky legal issue?
The Rolling Stones song Hey you, get off of my cloud was rather prophetic to the issue. Without throwing away the benefits that the cloud can bring is the development of the private cloud concept; that is, run smaller scale, non-global clouds tailored to local conditions. For example, a private cloud guaranteeing UK data protection at all times as part of the service level agreement to customers. Some larger organisations already run their own private cloud services and technology is now available that allows them to do this will relative ease (i.e. virtualization). This provides them most of the benefits of cloud with less risk. The offset however, is that you still need to manage your own servers which is costly. Remember that by owning the servers you are in full control of the data. For smaller organisations that don’t have the ability to host their own services the image is less clear. It could be that ‘localized’ application service providers (ASPs) will work within guaranteed data jurisdictions meaning that data is always legally protected. ASPs could form from internet service providers that would hold the key to success. They understand the context of UK Data Protection regulations already and so will have a handle on how to manage your data via software tools they will rent to you. It is unclear if or when such services will be developed and deployed however. It will also require that re-sellable software becomes available and at this time few exist outside of a few collaboration (e.g. Microsoft SharePoint) and customer relationship (e.g. Sugar CRM) packages. To be honest, cloud technologies are presently few and far between and the long-term viability of the cloud concept will only be sustained if more services become available over the coming years. We will also have to watch out for evolutions of the cloud. For instance, Google’s Chrome operating system is technically a hybrid of the cloud vision by allowing users to directly integrate with Google cloud. Given Google’s successful development and deployment of the Android handset technology this may be a space to watch out for. In mean time, we’ll just have to wait to see how the sky looks when the sun eventually rises. Only then will we be able to see how many clouds are on the horizon, global or localized. Or indeed, who will be prepared to inhabit them…
Monday 23 January 2012
New e-Research book published - Sakai discussion included
It's been a long time since I've had the opportunity to publish to my blog. The good news is that Rob Allan (STFC Daresbury), Rob Crouchley (Lancaster University) and Alastair Robertson (Me, Isle of Man College) published a chapter within it. This work is the culmination of our research into the applicability of Sakai as an e-research toolset. The title of the book is Collaborative and Distributed E-Research: Innovations in Technologies, Strategies and Applications
and the title for our chapter is Collaborative and Distributed Innovation and Research in Business Activity.
The book contents can be viewed at the IGI Publishing @ http://www.igi-global.com/book/collaborative-distributed-research/58272.
If you need to contact me regarding this chapter, or any other element of my research, email me at alastair.w.robertson_at_btinternet.com.
Happy reading!
Ali
and the title for our chapter is Collaborative and Distributed Innovation and Research in Business Activity.
The book contents can be viewed at the IGI Publishing @ http://www.igi-global.com/book/collaborative-distributed-research/58272.
If you need to contact me regarding this chapter, or any other element of my research, email me at alastair.w.robertson_at_btinternet.com.
Happy reading!
Ali
Tuesday 21 June 2011
Developing and managing online projects and collaborations
This blog was developed for those people that work in project teams but realize that new collaboration technologies exist that could improve productivity. The application of ICT is fraught with choice and many people are forced to make technology choice decisions when they are potentially under-skilled to do so. Although many people understand that there is a role for online collaboration environments in projects, few project leaders or managers have sufficient ICT know-how on the role, selection and application of them in their work. It may seem strange for technologists (i.e. the author) to be more concerned about the human element of project teams than the technology itself, but it isn’t. Project teams rely on solid and managed interactions within the team and also for those resources outside of the team (e.g. project partners). This has always been the case, and as past research shows, the human element has often been missed which has led to failed project collaborations.
There are numerous approaches that can be applied to understand the dimension of online collaboration development. Use cases force the would-be developer to look at the requirements of the collaboration system. Use cases are written pieces about a human process in the project (e.g. workflow or communication among team participants) that highlights the human condition needing to be fixed by the collaboration environment. By drawing up use cases, it forces you to think about the processes that are presently used by your people and also helps you identify how collaboration tools could benefit them. There are a considerable number of tools available to support online collaboration (e.g. forums, chat tools, document stores and video conferencing), but some may suit your project team when others do not.
There are several keys to successful deployment of any ICT based technology but none more important than understanding the needs of your people in the work you’d like them to accomplish. Teams are about harnessing knowledge that they then turn into novel innovations. Whether the innovation is an idea, concept or a produced product, collaboration environments can make readily knowledge readily available to your people and help them work more effectively. How you move from A to B is a critical question. A is the starting point on wanting to learn about deploying collaboration tools into your team environment, as evidenced by reading of this blog, B is your intended destination where you have functioning tools in use. Travelling from A to B can be a challenging project in itself, but nothing is more rewarding than watching your people produce output in front of your eyes!
People thrive on collaboration. Collaboration is the bringing together of minds to solve problems of the day, to provide a platform for developing novel innovations. People have been working collaboratively since man first began hunting animals in a team-like way; perhaps longer. Many example of humankind's collaborative ways exist. For example, during the industrial revolution new approaches to the production of harder metals took place by people travelling to other countries and seeing for their own eyes how other cultures had dealt with pre-existing problems. Word of mouth was key to their ability to determine if processes were available elsewhere they could take advantage from. Innovators take their lead from current practice then tweak it with their own knowledge, until something even better emerges. The process of collaboration allows innovation to occur and it should be actively encouraged. The human race learns from itself, from both mistakes and wins. Although today’s cataloguing of web material makes it much simpler to find out how ‘stuff’ is done, and who produces innovative knowledge; Google is about quickening resource discovery, but it still takes thought intervention to create novel knowledge from it. Google is simply a service platform that quickens our information find and publishing capabilities. Collaboration is about providing a people orientated platform where like-minded people of similar interest, but with unique specialism’s, can share knowledge to create a whole that is greater than the sum of its parts.
There are numerous approaches that can be applied to understand the dimension of online collaboration development. Use cases force the would-be developer to look at the requirements of the collaboration system. Use cases are written pieces about a human process in the project (e.g. workflow or communication among team participants) that highlights the human condition needing to be fixed by the collaboration environment. By drawing up use cases, it forces you to think about the processes that are presently used by your people and also helps you identify how collaboration tools could benefit them. There are a considerable number of tools available to support online collaboration (e.g. forums, chat tools, document stores and video conferencing), but some may suit your project team when others do not.
There are several keys to successful deployment of any ICT based technology but none more important than understanding the needs of your people in the work you’d like them to accomplish. Teams are about harnessing knowledge that they then turn into novel innovations. Whether the innovation is an idea, concept or a produced product, collaboration environments can make readily knowledge readily available to your people and help them work more effectively. How you move from A to B is a critical question. A is the starting point on wanting to learn about deploying collaboration tools into your team environment, as evidenced by reading of this blog, B is your intended destination where you have functioning tools in use. Travelling from A to B can be a challenging project in itself, but nothing is more rewarding than watching your people produce output in front of your eyes!
People thrive on collaboration. Collaboration is the bringing together of minds to solve problems of the day, to provide a platform for developing novel innovations. People have been working collaboratively since man first began hunting animals in a team-like way; perhaps longer. Many example of humankind's collaborative ways exist. For example, during the industrial revolution new approaches to the production of harder metals took place by people travelling to other countries and seeing for their own eyes how other cultures had dealt with pre-existing problems. Word of mouth was key to their ability to determine if processes were available elsewhere they could take advantage from. Innovators take their lead from current practice then tweak it with their own knowledge, until something even better emerges. The process of collaboration allows innovation to occur and it should be actively encouraged. The human race learns from itself, from both mistakes and wins. Although today’s cataloguing of web material makes it much simpler to find out how ‘stuff’ is done, and who produces innovative knowledge; Google is about quickening resource discovery, but it still takes thought intervention to create novel knowledge from it. Google is simply a service platform that quickens our information find and publishing capabilities. Collaboration is about providing a people orientated platform where like-minded people of similar interest, but with unique specialism’s, can share knowledge to create a whole that is greater than the sum of its parts.
Tuesday 15 February 2011
Six months of changes in Sakai...
Six months can be a long time in technology markets, and I have created this blog to provide you insite on changes we have made to out Sakai service. Only six month ago we moved over to Sakai 2.7. Bringing in new social networking features, it opened a door to new ways of thinking that make Sakai a warmer place to work in. By allowing Sakai users to form Connections among work colleagues (a bit like friends in Facebook) it allows groups of similar minded individuals to form up without the requirement of a worksite. The standard social networking kit in 2.7 is pretty basic and although people hooked up quite quickly, it soon became apparent that something else was needed to help it make better sense.
Adrian Fish had already begun developing the floating chat tool that would take advantage of Connections and it was launched just last month. This tool allows users to chat with Connections that are already logged into Sakai. It's been a great hit with our users. We noticed that even in the wee hours of the morning that people were still hooked into Sakai chatting away. We think this tool is an essential ingrediant to any collaboration platform, even for asking simple questions like, 'Time for a coffee?' or 'Have you seen this link?'. Not content with getting the core chat running, Adrian then set about some newer features, such as the ability to 'ping' connections that are not already logged into Sakai. This feature emails recipients an invite to chat with you providing them an additional opportunity to stay in the loop. We think that floating chat will become available in Sakai 2.9, but stuff doesn't always go to plan :(
Then Dan Robinson and Stevo Swinsberg decided to expand out the Profile2 feature set. Building a place to post to your own or your connections profiles, this is a space to share information among your social sakai contacts; we only launched it 10 days ago so we are still looking at user issues. To date, nothing dramatic has emerged and that is usually a good sign. The current publishing space is a bit limited in scope and we'll be adding new features throughout the year based on user requests. Dan also tidied up the people search facility within Profile so that presentation of results is much better making it easier read.
So, in six months we went from Sakai CLE to Sakai CLE with social connectivity. Last year some of my eBusiness students complained that Sakai was hard to get into (being the second VLE at Lancaster); since deploying 'boosted' Sakai users tell me they really like the experience for the additional functions we have added. We note also that the addition of social tools has improved Sakai 'Stickiness'; a term used to describe how long people stay logged in for. I guess this is what we should expect. Importantly also, due to implementation of Profile2 with our additional features, our latest clients tell us that profile2 was one of the strongest choice factors for choosing to use our technology. Moreover, discussions are now underway with central uni services to roll out Sakai across campus later this year. Watch this space!
During the next 6 months we'll be evaluating user data in various ways. For example, we have collected Connection formation data and can link this aggregately to other data arrays (e.g. propensity to chat, complete profile, activity in worksites etc) plus user survey data. With this information we'll be able to determine segments of user's and track back user acceptance issues from it. After all, the new ways of working won't be to everyone's taste, but it'd be nice to be able to find out who they are and tailor a service about them too.
Adrian Fish had already begun developing the floating chat tool that would take advantage of Connections and it was launched just last month. This tool allows users to chat with Connections that are already logged into Sakai. It's been a great hit with our users. We noticed that even in the wee hours of the morning that people were still hooked into Sakai chatting away. We think this tool is an essential ingrediant to any collaboration platform, even for asking simple questions like, 'Time for a coffee?' or 'Have you seen this link?'. Not content with getting the core chat running, Adrian then set about some newer features, such as the ability to 'ping' connections that are not already logged into Sakai. This feature emails recipients an invite to chat with you providing them an additional opportunity to stay in the loop. We think that floating chat will become available in Sakai 2.9, but stuff doesn't always go to plan :(
Then Dan Robinson and Stevo Swinsberg decided to expand out the Profile2 feature set. Building a place to post to your own or your connections profiles, this is a space to share information among your social sakai contacts; we only launched it 10 days ago so we are still looking at user issues. To date, nothing dramatic has emerged and that is usually a good sign. The current publishing space is a bit limited in scope and we'll be adding new features throughout the year based on user requests. Dan also tidied up the people search facility within Profile so that presentation of results is much better making it easier read.
So, in six months we went from Sakai CLE to Sakai CLE with social connectivity. Last year some of my eBusiness students complained that Sakai was hard to get into (being the second VLE at Lancaster); since deploying 'boosted' Sakai users tell me they really like the experience for the additional functions we have added. We note also that the addition of social tools has improved Sakai 'Stickiness'; a term used to describe how long people stay logged in for. I guess this is what we should expect. Importantly also, due to implementation of Profile2 with our additional features, our latest clients tell us that profile2 was one of the strongest choice factors for choosing to use our technology. Moreover, discussions are now underway with central uni services to roll out Sakai across campus later this year. Watch this space!
During the next 6 months we'll be evaluating user data in various ways. For example, we have collected Connection formation data and can link this aggregately to other data arrays (e.g. propensity to chat, complete profile, activity in worksites etc) plus user survey data. With this information we'll be able to determine segments of user's and track back user acceptance issues from it. After all, the new ways of working won't be to everyone's taste, but it'd be nice to be able to find out who they are and tailor a service about them too.
Friday 24 December 2010
What happens when the cloud bursts? Skype shows the risks...
I beleive that the recent failure of Skype shows how dependant some businesses are becoming on cloud based services. Although Skype isn't a true Cloud technology, as you need to install its software locally, the service uses the WWW in a 'cloud like way' to transport data. Skype's recent system failure (click here for BBC news story) have left businesses that are reliant on its services wide open to problems; especially those that have no alternative solutions in place. There are likely to be many more events like this over the coming years and the lessons learnt for some organisations may be terminal. The Cloud places risks into any adopting organisation and each risk needs to be evaluated with mitigaton strategies in place should specific 'critical' events occur. I question how many organisations, particularly smaller businesses, undertake this level of contingency planning however.
Subscribe to:
Posts (Atom)
