Last.fm and the application of the lightweight models and cost-effective scalability

Standard

Lightweight model and cost-effective scalability is a pattern that Tim O’Reilly identified as part of Web 2.0 platforms endeavours (O’Reilly, 2005, pag. 4). This pattern consist of maintaining the ability to scale quickly when required. The lightweight model implies that is not only a matter on technology, the business model of the company should be able to dinamically cater for a increasing number of user/customers.

Nowadays, cloud computing allows to rent utility computing services such as storage and computing resources tailored to the demands of the users; that’s why is affordable for Web 2.0 start-ups to rent those services on the Internet, instead of buying expensive hardware and software for its functioning. In the event of success and acquisition of a considerable user-base, the Web 2.0 platform would only need to upgrade their services in the cloud, so it could cater for more users. Facebook was based in this pattern, as was created by a few students with a small capital in Harvard University; then, as more users were adopting the platform more economical and technical resources were added to provide the needed increased power to the service.

Image

Let’s explain this pattern better with its relation to a popular music website: last.fm. Last.fm is a music platform that uses a recommendation system called Audioscrobbler, which helps to recognise the musical tastes of user based on the tracks that they listen to, either on online radios, user’s computer and portable music devices; this information is uploaded to a database via the music player or a plug-in (Wikipedia, n.d, para. 1) where is catalogued and used to make recommendations to users depending on their tastes.

Image

Starting small and growing from there: at the beginning Audioscrobbler was a university project of Richard Jones when he attended the University of Southampton in the UK. He developed the firsts plug-ins and created and API for the community to contribute to extend them. On the other hand, last.fm was a music site and online radio. Both merged to become a stronger company. But their beginnings were humble, start-ups with few resources making use of open source software in the case of Audioscrobbler (Wikipedia, n.d, para. 4-5).

Growing and acquisition by CBS – scaling up: By 2007 CBS, the giant american media corporation acquired last.fm for US $280m, the largest-ever UK Web 2.0 acquisition (BBC news, 2007). By this time it had been a long walk since their humble start-up and so, their user-base increased exponentially and of course their needed resources were much more, so using cloud computing to scale was the way to go to gather more storage and computing resources.

Image

Shoemaker to your shoe: For the creation of a lightweight business model is important to consider in which aspects the platform excels and what other aspects could be outsourced to business partners; linking different business models in synergy. With this approach the platform could focus in their competitive advantage and let other contribute in what they do best. In last.fm we can find examples of this: before, last.fm had its own online radio service; but realising that there were many other online radio stations and streaming services like Spotify and YouTube that were offering a better service, last.fm decided to cut their online radio and outsource it to those providers (last.hq, 2014). As we say in spanish: “Zapatero a tu zapato” – “Shoemaker to your shoe”, meaning that anyone should focus in their field of expertise and outsource other activities to companies with better expertise. In the case of Last.fm, their focus is the music recommendation system (Audioscrobbler).

Image

In a nutshell, the pattern is: start small, increase customer base, get more economical resources, scale up gathering more technical resources, outsource activities that are not part of the competitive advantage and grow, grow, grow but always making sure that the expansion of the company is proportional to its customer base.

Thanks for stoppping by and feel free to have your say!

 

References:

BBC news (2007). Music site last.fm bought by CBS. Retrieved from http://news.bbc.co.uk/2/hi/technology/6701863.stm

last.hq (2014). Did someone say on demand?. Retrieved from http://blog.last.fm/2014/01/29/did-someone-say-on-demand

O’Reilly, T. (2005). What is Web 2.0 – Design Patterns and Business Models for the Next Generation of Software.Retrieved from http://oreilly.com/pub/a/web2/archive/what-is-web-20.html?page=4

Wikipedia (n.d.). Last.fm entry. Retrieved from http://en.wikipedia.org/wiki/Last.fm

Advertisements

17 thoughts on “Last.fm and the application of the lightweight models and cost-effective scalability

  1. Hi Jacques,

    That was a great post about Last.fm. I have never personally used Last.FM, but it is interesting to see how Last.fm correlatively portrays O’Reilly’s 8th core pattern of lightweight models and cost effective scalability. Firstly looking at the origins of Last.fm, it is amazing to see how its overall demand and website has a large scalability, and how the company focuses on its internal development and how it has outsourced certain aspects of system to continue a lightweight model.

    Also, in addition to what you have stated there is also a business model evident within the website as the company employs paid subscriptions and advertising, much like Google AdSense to generate a revenue. Overall it was a fascinating read, and thanks for that!

    ~Jan

    • Hi Jan! I agree with you, it is usual to see the O’Reilly patterns show up in the Web 2.0 platforms, as Jason has said in the lectures: these patterns are still current and that’s why they have lasted so long in the Web 2.0 discourse. Last.fm has been built from the bottom as most of the other social media platforms, and that’s why there is the compelling tendency to be lightweight.

      Thanks for reminding the topic of advertising and paid subscriptions, I didn’t talk about it, but is an essential part of the business model. Both of these are very scalable through software, there is no need to have a lot of people overseeing or doing the job, just automated systems could be in charge of this. Definitely, is part of the lightweight and scalability features of Web 2.0.

      Thanks a lot!

  2. Hi Jacques,

    Short, sweet and to the point article. It is interesting to note so many companies had their origins in university computer labs, garages and dorm rooms by fresh graduates (and in some cases non graduates). I believe that in most cases starting small was a necessity and not an option owing to the lack of funds by the developers. It makes more sense if you see that facebook, last.fm, wikipedia and even major OS’s like Windows and Linux started as experiments where people just started something small and built their way up.

    las.fm dropping their radio service was something that I did not like, but it did make more sense for them to drop it. As they say try and try again, then try something else.

  3. Alejandro

    A VERY informative article, the information also backed with relevant up to date examples that we can all relate to. I agree with the previous comments, its great to see how small projects have grown to such huge proportions, university students, with minimal resources but ambitious, most importantly, its great to see that they adapted, making the relevant changes they saw fit in order to grow/change and move towards their targets and goals..
    Thanks for the great article Jacques

  4. This is a very interesting article that shows how small ideas can become great ideas! There are thousands of web services that you can be now easily customise and integrate to your service through their API documentation. It will allow your service/business to become stronger as you grow over time. The cloud computing is bringing new business models and changing the way people do business. Crowdsourcing, crowd-funding, the social enterprise, freemium models are all allowing individuals or small businesses to fund their projects, expand their services and to contribute to the community.

    Thanks for sharing these inspiring examples 🙂

    • Thanks for your support Angelo! I agree, the power that gives the integration of relevant web services is pretty intense; how to create mash-ups from different sources has become an art for the developer and an important way to reach audiences innovating in assembly. Cheers!

  5. Very interesting, I found many of the artists that I hear today using last.fm recommender system…

    Ironically, once they shut down their radio service, I stopped using them, I tried to install the plug-ins to record my musical taste and keep the recommendations, but I didn’t find good plug-ins… and also I stopped going to the site, I don’t check my last.fm recommendations very often, there’s nothing that reminds me of doing that… maybe if they sent emails with suggestion every once in a while, or if they connected their service to facebook and integrated the music “likes” of people, or suggested friends in facebook (Twitter, Google+) by their musical taste or something similar, I would still be an active user…

    • Hey Sebas,

      Thanks for your visit. You touched really good points there, I like particularly the one that last.fm could integrate with Facebook and suggest friends with similar musical tastes.

      I didn’t know last.fm before this post, so I don’t know how was the radio service; what was so special about it?

      Cheers!

      • It was one of the first radio services I knew that recommended music based on your musical taste, it used AI to do it, like Amazon does to recommend you products that you would like, based on the products you have searched and bought.

        So, the recommendations were pretty good, I liked almost every recommendation I received there, based on what I was already listening to. And they recommended a band and then you could listen to it immediately, you could just listen to your personalized “radio”, based on what you had already listened. And they also kept track of the songs you heard and the songs you skipped to improve further the recommendations.

  6. Thank you Jacquesker for writing another great article. I will definitely remember the saying “Zapatero a tu zapato” from your article. I do have one question I would like to ask though. Do you think it is a wise move to outsource hosting of Web 2.0 applications to the cloud. According to Price Waterhouse Coopers (pwc, n.d. ), one of the main advantages of the cloud it that it’s possible to be scaled quickly and at a fraction the cost?
    From my experience with using Microsoft Azure cloud it is possible to allocate multiple websites to a single resource pool, which will scale the number of servers being used based on combined usage. This does allow multiple web applications to be scaled in a cost-effective manner as you discussed in your article. Do you think that this method of cost-effective scalability will become widely accepted in future web applications?

    References
    Pwc. (n.d.). Cloud computing gets strategic: Reducing technology costs is just the starting point. Retrieved May 17, 2014, from http://www.pwc.com/us/en/view/issue-13/cloud-computing-gets-strategic.jhtml

  7. Thanks Jacques, for an awesome read, nice, concise and jam packed with info. Loved it.

    I really enjoyed last.fm a couple of years ago before the other online radios really took off. Their song recommendation system really is a great example of the application of lightweight models and cost effective scalability. By making a conscious decision to cull their radio service they have maintained and ensured their viability in the marketplace.

    Do you think there are any other services currently available that could benefit from reducing their provided service to focus solely on their prime directive?

    Sam

  8. Hi Jacques,
    I used to be a big fan of last.fm in my early teenage years, so I found it quite interesting to see how this website has evolved since I last used it. You’ve presented all the information in regards to lightweight models and cost effective scalability brilliantly, highlighting how this platform caters for large audiences, personalising suggestions for each user based off their individual music preferences.

    Taking on an online radio capability was the next logical move for last.fm, scaling up and expanding its services to capture a larger market. I think it was a very smart move for last.fm to outsource their online radio, allowing for the company to bring their focus back to what makes them individual and popular in their own right, as the market giants (YouTube and Spotify) offer superior online radio services and have a strong capitalisation of the target audience. I haven’t got a lot of experience with Spotify, but its popularity with the youth of today is undeniable, making it particularly difficult to infiltrate due to its partnership with Facebook (Simpson, 2011).

    Last.fm’s music recommendation system was what made me become a huge fan initially, so I’m happy to see they’re going back to their roots and nurturing their niche area. Discovering new music that meets personal tastes is an art, which, from experience, last.fm does very well! I hope they embrace what makes them unique and powerful, before taking on new projects, as I know they have such great potential to go far in the future of music.

    Aside from online radio, perhaps future endeavours could include facilitating local bands getting discovered to the community rather than already-established bands; or even becoming affiliated with musicians and record labels, developing into the platform of choice for discovering upcoming gigs, ‘meets and greets’, and releasing new music first.

    Music is such a core element in life; the future is bright and has so much opportunity. Thanks for the great read!
    – Steph

    References:
    Last.fm. (2013). About Last.fm. Retrieved from http://www.last.fm/about
    Simpson, P.V. (2011). Spotify and Facebook in music partnership. Retrieved from http://www.thelocal.se/20110923/36316
    Spotify AB. (2014). Features. Retrieved from https://www.spotify.com/au/

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s