-Recognition of the critical role of software in promoting and maintaining sustainability has been emerging recently. Yet, over the last 60 years a vast amount of software has been developed and deployed without any consideration of sustainability. This paper presents a case study on how relevant sustainability requirements have been identified (using goal-oriented techniques) and integrated into an existing software design and implementation. It also discusses metrics used for evaluation of the sustainability effects of the new design. The evaluation of this study demonstrates that the re-designed software is both more extendable and more sustainable. Thus, this paper presents an example process and case study for undertaking sustainability-centred software re-design, as well as reports on a number of trade-offs and interdependencies between the studied sustainability requirements.
While integration of ICT into business models and
processes has a well-documented economic benefit, the
environmental and social aspects of its’ impact are less
clear-cut. For instance, while some e-businesses, such
as eBay, encourage reuse of physical goods, they can
also cause over consumption [
In this paper we present a case study of such
requirements and the effects they have on an e-shop business
for selling pets (called PetShop). The aim of this study
is in determining how retrofitting sustainability into an
existent system will affect that system. Following [
We also present a process for identification of
sustainability-related requirements and evaluation
methods and metrics for some aspects of the economic and
environmental sustainability requirements identified in
this case study. Throughout this study we use the relevant
consolidated research on sustainability undertaken so far
in the software engineering community and extend it
where open issues are identified. In particular (similar
to [
Since the present study aims at consolidating the relevant research of software engineering for sustainability, we must start by reviewing this research:
Naumann et al. [
Cabot et al. [
The Yellow project [
We consider [
The idea that e-business is conducive to sustainability
is also advocated in [
In order to measure the impact of sustainabilityrelated requirements of the engineered system, one needs to utilize some evaluation metrics. Here two distinct perspectives can be adopted: one focusing on the sustainability of software itself (i.e., software quality), the other on the impact that this software has on the broader environment (ecologic, economic, and social) within which it is situated. The prime interest of this study is in broader impact of the software system.The present study aims to utilise the pre-existing metrics to the highest possible degree, only providing additional metrics where, to the best of our knowledge, no previous metric exists.
When evaluating the impact of the software on the
business and its wider environment, the related work
suggests focusing on the system effects in terms of
energy consumption or CO2 emission. For instance, [
Thus, the contributions of this paper can be summarised as: 1) presentation of an end-to-end re-engineering study where sustainability requirements are identified, implemented for a pre-existing system, and evaluated; 2) presentation of a set of metrics suitable for quantification of environmental sustainability in requirements; 3) demonstration of the measurement process of environmental sustainability requirements of a software system; 4) demonstration of the application of the consolidated research on software engineering for sustainability.
Next, Section II of this paper presents the Pet Shop case study. It is followed by discussion of the methods and metrics used in this study in Section III. The sustainability requirements and their mutual dependencies and contributions to environmental impact are discussed in Section IV. The details on the actual and expected impact due to system re-engineering are presented in Section V. We then discuss several points relevant to this study in Section VI.
Java Pet Store 2.0 is a Reference Application [
The Pet Store application provides basic e-shop functionality without concerning the sustainability requirements. The supported functionality is shown with white and yellow coloured use cases in Figure 1 (use cases in red are to be discussed later). The Pet Store documentation states that:
Seller can upload information on a pet (including its name, description, price, image and seller’s contact information) to a specific category. The seller’s contact requires last name, first name and address. An image thumbnail will be created corresponding to the uploaded image; Seller can tag an item; Web user can locate items by choosing category or searching them by tags, location or keywords; When viewing an item, web user can also rate it in range from 1-5.
Web user can set an item as inappropriate;
The application can get and show RSS news of Java Blueprint on the top of the page; The application assumes that there is an existing list of categories with their appropriate images such as “Cats”, “Dogs”, etc. Besides, each category has a list of products with their appropriate image such as “Medium Dogs”, “Small Dogs”, etc. An item is categorised according to products.
As shown in Figure 1, three types of actors are distinguished:
A viewer, who is an ordinary web user and can access the PetStore site; A member is a viewer who has signed up and logged in into the system; A manager administrates the PetStore application. Manager has some unique behaviour, e.g., creating and viewing reports (as defined in Figure 1) but also has the same access and use rights as a member.
The colour of use cases in Figure 1 is used to distinguish the originally defined and modified use cases: use cases which have been defined in the original system and remained unchanged throughout the below discussed system re-design are not coloured (i.e., are white); the yellow use cases were defined in the original use case model, but their behaviours had to be changed due to the newly defined activities/requirements for improved sustainability; the red cases are the newly defined use cases, that are to be discussed in following sub-sections.
In order to undertake the intended system re-design study, we must define a process for relevant requirements identification as well as a mechanism for comparing the updated and original versions of the system. These are outlined in the following sub-sections.
As discussed previously, a number of researchers ([
Using these generic goal structures as starting point, we instantiated them within the context of the PetStore application and operationalised them into tasks. These tasks (also shown as red and yellow use cases in Figure 1) have then been integrated into the updated system design and implementation.
To represent the goal decomposition and contribution
models, we use the i* language to capture the context
as well as relationships between sustainability and other
system goals [
As pointed out above, our prime interest is in studying how consideration of sustainability requirements in a software system impacts the sustainability, i.e., the broader (ecologic, economic, social) environment within which the software system is located. Thus, we are interested in measuring environmental impact (i.e., impact achieved due to use of software) in terms of emissions.
mental impact of the re-design of the system can be measured by comparing its CO2 emissions to emissions from the original system.
In particular, it can be estimated using the two following facts:
1) Fact 1: The amount of CO2 emitted (CO2e) for
a unit of electricity from UK grid2 is about 600g [
2) Fact 2: Empirical measurement of power use with various types of servers has demonstrated a stable linear 1We used UML for later stages or detailed design.
2It is worth noting that this emissions number is very
countryspecific, ranging from 60g. for Iceland to 1060g for Australia. The
exact value depends on the fuel mix used by the given country’s
grid with Iceland using only natural geo-thermal sources where other
countries use a mix of sources, such as fossil fuels (oil, gas, and coal),
nuclear power, renewable, etc.
relation (accurate to within 5%) across all CPU
utilization rates [
Px = Pmin + x*(Pmax-Pmin)/100 Where : Px: Power consumption for x%CPU (watt) Pmin : energy consumption in idle state (watt) Pmax: energy consumption for 100% CPU usage (watt)
For example (as per [
Thus, the [x*(Pmax-Pmin)/100] component in the above equation marks out the additional energy used for the completion of a given task (on top of the idle state consumption). Since, as per Fact 1, each unit of electricity from the UK grid produces 600g of CO2e , the CO2e for a task (COx) can be calculated in kilograms per Watt as:
COx = 0.6 * (Px - Pmin) = 0.6 * x*(Pmax-Pmin)/100 = 0.006*(Pmax-Pmin) *x
To account for the duration of a given task, this formula needs to be further adjusted for the time. Thus, the emissions (in kilograms per Watt) for a task with a given duration in seconds (Tx) will be calculated as: Tx = COx*3600/ Caverage seconds Where:
Caverage is the average CO2 emission (CO2e) in kilograms per hour
COx is CO2e (in kilograms per Watt) for doing the task.
work: In order to utilise the above discussed metrics, the study needs to collect the relevant data. Data for environmental sustainability requirements is to be evaluated based on CO2 emission and task processing time. CO2 emission for a task is measured as Cx = 0.006*(PmaxPmin)*x. And the appropriate processing time for a task is approximately Cexp*3600/ Caverage seconds in which Cexp is the expected CO2 emission and Caverage is defined according to testing system.
Therefore, to collect data for measurement of per-task
emissions, CPU usage on server side and processing time
on client side should be monitor throughout processing
of a given task. The data collection set up is illustrated in
Figure 2. For the present case study, a Java application
was implemented using Sigar library [
On the client side, we use the Firebug [
Fig. 2. Data collection setup for Pet Shop
We use black box testing to verify the implementation of our re-design for sustainability by defining a test scenario for a task to be evaluated, recording the Cx and Tx for each task in both original and updated versions, and comparing these. Where a task to be evaluated is not supported by the original version of the system, the data on updated system is compared against an impact value obtained from an alternative buy comparable way of conducting the required task without the system support (further discussed for each relevant task in section IV).
IV. SUSTAINABILITY GOAL ANALYSIS FOR PET
SHOP APPLICATION
Having outlined the goal-based analysis process and metrics to be used for case study evaluation, we now present the decomposition and operationalisation analysis of sustainability goals. Since the combination of all goal models on sustainability generates quite a large and complex picture of requirements, below we introduce the three sustainability views one at a time. For each of the discussed views the task operationalisations are also discussed in terms of their expected environmental impact.
In economic sustainability view the emphasis is on two main goals: increasing profit and reducing cost.
Figure 3 demonstrates the decomposition and operationalisation of these two goals.
For e-business such as PetStore application, one of the most important factors supporting its’ economic sustainability is the number of users accessing the website (shown as “Increase Number of Users” goal in Figure 4).
In other words, the more users use the website, the more
potential benefits it will achieve. In accordance with [
Manage offers task allows sellers to view/add/remove/edit their uploaded pet items. The current system allows only uploading information, without support for further review and change to the uploaded content. This is an obvious shortcoming in terms of system usability for clear information presentation to the seller. With the additional functionality to mange the sale items, a seller can track the status of his/her items, change and upkeep sale details. This also helps to reduce website support cost: instead of the seller sending change requests to the site administrator, he/she can make the necessary changes by him/herself. The contribution chain of this task is depicted in Figure 4.
Sustainability impact measurement: In the original version of the system, where this option was absent, whenever a seller needed to make a change (e.g., change name or price of the offered item), he/she had to send an email to the system requests, asking the administrator to make the change. After receiving the email, the administrator would browse for the specified item and apply the requested change. Therefore, to improve system’s impact, the CO2e for making a change (Cx) with the updated version of the system should be less than the sum of CO2e caused by a i) sending a change request email and ii) carrying out the change by the admin.
According to Berners-Lee [
Let’s assume (for simplicity) that the only difference
in handling the change in the original and updated
systems are in sending email for administrator to make the
change vs. doing a web search to implement the change
in person by the seller (after which both sides will
presumably use the same web interface to carry out the
change). Then, to improve the system’s environmental
impact, Cx for seller should be less than 3.6g (i.e., 4g.
i.e. (0.05g+0.35g)). Moreover, as an average user takes
about 30 seconds to search, wait and scan the result from
the web search [
Manage orders task allows buyer to view/add/remove/buy pet items. With the original Tx should be less than spent while using the original version.
Add guideline task assists customers in understanding how to use the web application. The guidelines are provided by the site Manager. This task helps to ease the site use via clear information presentation, which is conducive of user number increase. The task contributions are shown in Figure 6. system a buyer can buy only one item per transaction.
The extended system will allow the buyer to reserve several items while shopping, and then process the payment for all reserved items at once. For multi-item Fig. 6. Task: Add guidelines purchases, this update will both reduce the need of repeated data exchange and payment processing between Sustainability impact measurement: this task will inthe customers and server, and make shopping process crease the environmental impact of the system by the easier. As shown in Figure 5 below, this change should value of CO2e for web search and review of the help result in reduced energy consumption (due to reduction pages, but will decrease the emissions by the value of in payment request processing). On the other side, we incorrect system use (e.g., mistakes in sale uploads and observe a negative contribution between the “increase related correction requests, etc.). The net impact value number of users” and the “decrease data transfer” goals. of this task is very dependent on the help content and However, this is unavoidable as attracting more users system use patterns. will lead to the need to process more data. This tradeoff Inform by email task informs the corresponding buymust be resolved through discussion with the owners of ers and sellers on the payment processing success or the system. failure. This is a necessary function to build the users’
This task can be operationalized via addition of “Add trust in the business’ service. The impact of the task to cart” and “Card” links to the system. is two-fold: it increases the perceived reliability of the
Add to Card allows buyers to add item to a cart. The web application, but also increases CO2e emissions due item is marked as “Booked”. to extra emails. The contribution chain for this task is Cart link allows buyer to view/remove/buy items. shown in Figure 7 below.
Sustainability impact measurement: the original ver- Sustainability impact measurement: this task will insion of the system supports only buying one item at one crease the environmental impact of the system by the time. To improve sustainability, the updated system must, value of CO2e of an email for each purchase. On the at least, deliver the same performance, as use of the cart other hand, the improved trust into the system by users will, at minimum, improve the shopping experience (i.e., is not easy to measure, especially in the present case social sustainability). This means that the CO2e and time study where the system is not in actual use. processing for buying a single item in updated system Online support task provides quick on-line help to must not be greater than that of the original version. the customers, improving their shopping experience. Moreover, the summative power and time consumption However, this will negatively affect the application perfor buying n items at once has to be less than that of formance due to use of real-time technology, which also buying n items one-by-one. This means that Cx should increases energy consumption. Besides, the real-time be less than one caused by the original version and support leads to additional costs required for continuous presence of an employee to provide the requested help.
The contributions for this goal are shown in Figure 8 below.
Add comments task allows users to add comments to each pet’s items or to news items. This is a mechanism for user-initiated discussions as well as mutual support/help between Pet Shop customers with regards to specific items. This will improve shopping experience, make the website more interactive, as well as more socially inclusive (and so, more socially sustainable).
On the other hand, the energy consumption for data exchange and storage will increase, as shown in Figure 9 below.
This task can be operationalised via addition of the “Comments” option that allows users to add/edit/remove their comments for each item.
Sustainability impact measurement: Although this
function increases CO2e due to posting on the web
site, it is also very useful in attracting more users
and supporting community. On the other hand, it can
indirectly support users’ query resolution. In absence
of comments, an email has to be sent for each query
(e.g., to administrators, or sellers) with the cost of 4g
CO2e per email, out of which 0.4g CO2e is caused by
the internet infrastructure [
Thus, Tx should be less than 30 seconds (which is that
spent for sending, receiving, and reading an email as per
[
For environmental sustainability we also adopt the
overall “reduce-reuse-recycle” perspective advocated by
Cabot et al.[
In context of PetStore application, reuse and recycling of other products made for pets and pet owners (e.g., carrier cases, accessories, etc.) were considered. Figure 10 presents the goal contributions identified for the environmental sustainability of the Pet Store, with a number of tasks operationalisations discussed below.
View distance and Find shortest path tasks aim at reducing pollution caused due to transportation. View distance shows the direct-line distance between user’s given location and the location of the item-to-be-purchased. Find shortest path provides travel directions for the route between these points. These functions are useful for both buyers and the Pet Shop business. The buyers may be choosing to purchase in environmentally-friendly way (with transportation emissions in mind), or planning on collecting items in person. The Pet Store will also use this for choosing their delivery and/or collection routes. The contribution structure of this task to the goals is shown in Figure 11.
This task can be operationalised via addition of the “View Distance” option.
causes about 0.7 CO2e which includes 0.1g for
searching, waiting and scanning results, Cx for this task should
be less than three searches, i.e., 3*0.6g = 1.8 g.[
Add category and Add product tasks allow for the PetStore to sell not only pets but also related pet products. In practice, when a user sells a pet, he/she often wants to sell the pet’s accessories as well. This functionality then helps to reuse/re-sell pet products. The original application supports selling pets only. Thus, this functionality helps in both reusing items (reducing emissions for new item production and improving environmental impact) and in extending the business to other related markets, which attracts more users and improves profits for the business, which is shown in Figure 12 below.
These tasks are operationalised by providing the “Add Category” and “Add Product” options for new categories and products respectively.
Manage donations task allows the seller to donate the pets and/or items to a particular charity. If this option is used, the Pet Store will receive payment from the buyer purchasing the donated item and transfer it to the charity selected by the seller. This will not only help with reusing pet items but also support the community at large.
This task is operationalised via “Donate” option which allows users to donate items to charity. Such items are marked as “donated”. The contribution structure of this task is shown in Figure 13.
Sustainability impact measurement: Because this is an additional choice option added on top of the already present functionality of uploading an item for sale, it is necessary to maintain or improve the performance of the original version. Besides, this function supports social sustainability, by promoting positive, helping attitude in the participating Pet Store user community. The social impact could be measured via responses from users over time, but that is not a part of the current study. For the purposes of this study we expect that Cx and Tx should be no higher than those of the original version.
Another utility of used items is in Recycling. Recyclable products can be donated for material reuse. Thus, recycling items can be listed by customers, collected and processed by recycling service providers such as factories. Pet Store will donate the profits from the recycling contributing to social well being, as well as improve environmental impact via reuse of materials. In general, although the energy consumption of the Pet Store system will increase due to more tasks to be processed, this task will reduce the overall pollution if it is well integrated with the “find shortest path” task discussed above. The contribution structure of this task is shown in Figure 13.
This task is operationalized via “Register recyclable items” option that allow users to upload recyclable pet items into relevant category (e.g., Recycle into pet products, corrugated cardboard, plastic bottles, hangers, bags) for donation.
ries out a transaction for donating used items of a given
weight (kg). According to [
In terms of social sustainability, we review the requirements that aim to achieve customer satisfaction and knowledge exchange in society. Both of these tasks help in contributing to long-term social well-building. The goal model for social impact is shown in Figure 14, with a set of relevant tasks analysed below.
Add news task allows site manager to add/edit/remove news. News are articles related to pets, such as pet policies, pet care or training programs.
This task is operationalised by provision of the “Add news” option to the site.
Sustainability impact measurement: This task must
cost less than a printed latter or an advertising email
which carry out the same function as posting news
articles. Since cost of a non-spam email is at 3.6g - to 50
g., excluding emission because of internet infrastructure
is lower than that of printed latter (140 g), CO2e for
processing this task should be less than sending an email.
As one may note, while in previous discussions we have
used the lower limit of the email-motivated emissions,
in here we would consider a somewhat higher value, as
news articles would normally be longer than a standard
email. It is known [
Share page task allows users to share pages on social networks such as Facebook, and Twitter. It not only helps the business reach potential users but also helps to share knowledge in interested communities, e.g., how to train a pet, or care for it.
Sustainability impact measurement: The effect of this task comprises of submitting a single ‘share page’ request to a social network site. Thus, CO2e caused by its processing must be less than of a single web search (i.e., 0.7g.) which would otherwise be needed (by, e.g., a Facebook user) for searching for and finding the information on a page of interest.
Support discussion: This task allows users to create and discuss specific topics. It gives a place for members to exchange knowledge. However, this task will require extra effort for managing and exchange of data, which can impact on performance of more prominent functions such as selling items.
Reporting allows managers to view sustainabilityrelated data, such as profits, donation rate, rates of selling recycleable and reuseable items, CO2e emissions, number of members at a given time, etc. Strictly speaking, this task did not originate in the goal analysis patters we chose to instantiate. Instead, it surfaced from re-design process. This task eases the managers’ work through coherent information provision.
vant information, CO2e must be less than those for using computer in 10 seconds, which is about how long it will take to submit a report request, get the results back, and quickly scan them. This means that Cx <Caverage/360g and Txe <10 (s).
To evaluate sustainability requirements, task simulation was set up on a laptop with the following configuration: Gateway Intel(R) Core(TM) i5 CPU M430 @ 2.27GHz, 4GB RAM. A server was deployed on Glassfish 3.1.2 run with Java EE 6.0 and built-in Derby database. On client side, the chosen browser was Firefox 20 because it supported the Firebug plugin which is used to collect data to measure the task processing times.
As discussed in section 3:
1) For calculating CO2 emission corresponding to
CPU usage, it is necessary to determine maximum
and minimum power usage of the computer used.
We used a power meter Energenie [
Another parameter used in this evaluation is time of
processing (Texp). In the scenarios supported in both
versions of the system, time of processing of new version
is compared with that of the original version. Where a
task is newly defined, and so absent from the original
system, Texp is defined according to Caverage and Cexp.
Caverage is amount of CO2 emitted by an efficient laptop
in an hour, which is set to 25g as per [
Thus, in order to demonstrate improvement in the system sustainability, the CO2 emissions (Cx) for the tasks defined in the updated version should be less than Cexp and the processing time (Texp) has to be less than Cexp*3600/Caverage seconds.
Table I shows the values collected from running a selection of the scenarios from the above discussed tasks. Scenarios used for evaluation of the updated system through simulation are those for which we could collect comparison data either from the original version, or by building a comparable estimate form alternative tasks. Since the present system is not in actual use, we had no access to users to conduct any social sustainability comparison, thus all such scenarios are left out. The summative results show in the Summary row of Table I demonstrate that the updated version of the system reduces overall emissions by about 25% and the overall required processing time by a factor of 28. These results are obtained with substantial new functionality added to the system.
The above results demonstrate that the actual values are not always within the bounds of the expected ones.
The TC01 scenario of the Edit an offer task has CO2e of 3.13g which, as expected, is less than that caused by requesting changes via email. Thus, this function helps users to save time and resources. Also, the time processing for this scenario is around 30% of expected time. Although the test is run over a local network this time efficiency suggests that even when running over internet, the processing time should take less than 30s.
In the TC02 scenario of Add item to cart task the buyers have not yet requested to buy an item, but have placed it into a cart, which gives them a change to review their decisions and expenses, without forcing the purchase. This is a positive contribution to the shopping experience. The amount of CO2e of this task is about 30% of the respective processing in the original version which is accessing Paypal service without processing any payment (estimated to 4.26 g). The time used for processing, however, is nearly double that of expected, which could be explained by the need for the application to access the database, mark the chosen items as booked, and return.
In scenario TC03 to buy 1 item through Cart the server processes the purchase of a booked item from the member’s cart. In the original version, this action directly connects to Paypal service (but does not carry out processing due to security concerns). In the new version it instead connects to the server to simulate payment. In terms of CO2e both versions are similar, with the updated system causing approximately 74% of the original one’s emissions, and so showing some improvement of application. However, it has a longer processing time (5 seconds above the original system’s). This is caused by the need to enter shipping address and create a payment (coming to 3.5 seconds), as well as the additional information provided in the new layout display, which takes additional time when a user browses the catalogue. Yet, it must be noted also that the original version doesn’t allow for simulation of an actual payment processing, so the time taken for this task in that version is likely to be under-estimated.
The TC04 scenario extends the previous one through Cart scenario with purchase of more than one item. We observe the amount of CO2e emissions is less than that of the original version. Moreover, there is only 9% increase (from 74% to 83%) when buying 5 items instead of 1. However, processing time is more than 4 times of the expected. Looking at this time in more detail, we note that it takes about 2 seconds between booking each item, while in the original version the in-between time is shorter because a new window is opened for processing purchasing, which was not taken into account in the previous measurement. Moreover, as in the case of scenario TC03, the original version does not support actual payment processing simulation, while the updated one does.
In TC05 scenario for Add a comment task the length of the comment is limited to 255 characters, which is comparable to an average (non-spam) email. The emissions and time used results include monitoring the whole process (i.e., signing in, posting and reloading comment list). Therefore, 50% reduction in emissions and large reduction in processing time (over 8 times) are good results.
The TC06 scenario of View distance task emits nearly exactly as much CO2e as 3 single searches set as expected value, with only 0.1 g. more. One of the main reasons for this extra bit of emissions is that the system connects to Google Maps service and downloads images which usually take longer to load than text. Moreover, this task also helps to improve social sustainability. For example, user now can visualise item’s location and Task Scenario and Steps TC01: Steps for Edit an offer: TC02: Steps for Add item to cart:
Logged in successfully with correct username and password Click Offers > Edit for an item Browse for a pet item Click “Add to cart” Repeat for original version: browse for a pet item then choose “Buy now” Logged in successfully with correct username and password Browse a pet item Click “Add to cart” > Click “Cart” link Enter shipping contact Choose “Buy now”
Repeat for original version: browse for a pet item then choose “Buy now” TC03: Steps for Cart for buy 1 item: Logged in successfully with correct username and password Browse for a pet item Click “Add to cart” > Click “Add to cart” Repeat above steps 5 times Enter shipping contact Choose “Buy now”
Repeat for original version: browse for a pet item then choose “Buy now”, do this for 5 items TC05: Steps for Add a comment:
Browse a pet item
Submit comment “This is TC05” TC06: Steps for View distance
Browse a pet item Enter zip code “94105” to “Your location”
Click “Get distance” TC07: Steps for Add news TC08: Steps for Support donations
Logged in successfully with correct username and password Click “Admin” > “News” tab > “Add” Enter required information for news Submit Browse the news Logged in successfully with correct username and password Click “Offer” > “Add new” Enter required information for a pet item Select “Donated” option Submit
Repeat for original version (without “Donated” option) TC09: Steps for Register recyclables
Click “Recyclable” Specify “Weight” is 1 kg and other required information
Submit TC10: Steps for Create report TC11: Steps for Load reports SUMMARY
Logged in successfully with correct username and password Click “Admin” > “Report” tab > “Create Report” Logged in successfully with correct username and password Click “Admin” > “Report” tab > “Create Report”
Select “From” and “To”
4.26(*) 4.5(*) estimate distance and cost of travel before making a decision on purchasing or transportation. It also improves awareness on CO2e emissions to environment caused by transportation, as well as reduces processing time by factor of about 76.
The TC07 scenario for Add news evaluates to very effective results with the emissions coming to under 50% of the expected value and over 100-fold improvement in processing time.
The TC08 scenario for Support donations task extends the original version (with marking selected items as donations) with additional slight reduction in resultant emissions. The processing time for uploading donation items is also twice as quick. The time and emissions are reduced since the user has to sign in with the updated version but must type in his/her contact information for every new submission with the original system (as it does not support user accounts).
While the TC09 scenario for Register recyclables task takes much less time than expected, it causes 2.99 g CO2e emissions - about 5 times greater than a single query (which we use as the expected emissions target). Yet, as discussed in previous session, the overall environmental benefit from this task is likely to be much greater due to reduction in virgin resource extraction demand. This, however, is not evaluated in the present study.
The TC10 Create report and TC11 Load reports scenarios for the Reporting task both have much higher emissions than the 10 seconds use of an efficient laptop. One reason for this is that to create a report the administrator has to access the admin page on the web and select the categories, products and articles which are listed on it. This takes effort from both server and client to process. However, the actual emissions values are rather small and the processing time is also short. Since, these functions support monitoring sustainability benefits, which helps the manager to evaluate and plan further actions, all of which is useful to a system.
Validity of Results: While all code and experimental data presented in this paper are freely available for crossvalidation, there are two issues that are worth pointing out with regards to the validity:
The estimates for CO2e emissions are largely taken
from [
When choosing an expected emissions level (in section 4), we have consistently looked for lowerbound estimates which the updated system would have to fit into. Thus, it is very likely that the carbon reduction from the presented redesign is in fact larger than what we have reported it to be.
Tensions within Sustainability: The findings from this case study demonstrate that there is an ever present tension between the three “pillars” of sustainability, i.e., its environmental, economic, and social sides. This is because economic sustainability for a business supported via software aims to increase consumption, while environmental sustainability aims to reduce it. The social sustainability, in turn, has a conflict with itself - aiming at both increased, faster, and more convenient consumption available to all and vibrant, un-spoiled natural environment at the same time. Yet, as it stands today, increased consumption inevitably lead to environmental depletion. This conflict has been observed within several discussed requirements (e.g., should the system provide continuous on-line support at increased economic expense? should the system increase number of users at the cost of increased emissions?, etc.). While conflicts within system goals/requirements are quite commonplace in all development projects, there is an additional diminution to sustainability conflicts: since social and environmental dimensions are “invisible” , until recently, the monetary returns have massively biased trade-off resolutions in favour of income maximisation. It is now necessary to educate software professionals on the value of the invisible environmental and social dimensions. This does not mean that businesses must always lose income. On the contrary, many environmentally or socially motivated requirements will turn a profit. For instance, in Pet Shop, registration for collecting recycled products, not only helps to reduce a large amount of landfill rubbish but also open new markets of providing materials to potential manufactures, which contributes to business profit.
Domain-Dependent Sustainability: It should be noted that, as each domain has its own requirements, so sustainability has different manifestations for different domains. Sustainability for each domain must be analysed with different criteria. PetShop application, for example, is examined mostly under supply chain context which focuses on how to increase the sales, effectively exploit existing resources and extend the market. Therefore, besides basic function of selling products, the indirect criteria are improving customer satisfaction, reusing or reducing resources through all phases of e-business from viewing products to processing orders. Goals like customer convenience, optimised travel routes, new markets etc. become important sustainability concerns for an e-business. This list will inevitably be very different for a different study domain.
Metrics and Measurements: As discussed above, we utilised measurements via power consumption and processing time to quantify environmental impact of the software. This worked well in most scenarios, however, in a few cases we observed a conflict between different runs of the scenarios. For instance, in case of TC09 2 runs of the scenario measured to a significantly lower time than the other 3. It may be because some tasks depend of several ‘outside’ factors, such as network communications paths, loads on the DNS servers, etc. This issue needs to be further investigated. Additional work is required in evaluating the impact of the redesign on the software quality through additional metrics. We also observed that measurement of social sustainability impact was not feasible in this case study as our case study system was not in real use. However, the measurement process and metrics for this area of sustainability are also grossly under-researched at present.
With continuous improvements in technology, more and more applications are developed to support human demands. However, when software is integrated into physical world, it affects many wider aspects of life (such as environment and society at large) which are very often not considered relevant during software development. This study reports on an experience of discovering and integrating such wider requirements into an existing application. Throughout the case study, we have observed a tension between different dimensions of sustainability. Thus, it is not sufficient to simply ‘increase’ sustainability of software, as this may very well mean that one sustainability dimension (e.g., economic) of that software system is very well supported, while others are neglected. For a system to be truly sustainable the increase in sustainability must also be achieved in balance, so that all pillars of sustainability are equally well supported. In the presented study, we have considered economists, environmentalist and socially concerned stakeholders, as we think that including such diverse stakeholders is a productive approach for capturing sustainability goals. Evaluation methods and metrics are also proposed for environmental and economic requirements. Here we convert performance of requirements into an amount of CO2 emissions, which is used as a common unit to measure sustainability. Overall, we have identified 19 sustainability requirements out of which which 11 were implemented, and the re-designed system was evaluated. Compared with the similar functions in the original version, the new version reduces CO2 emissions by 20% and increases processing time within the system by a factor of 28. As a future work, we are researching on factors that impact on e-commerce application in terms of the relationship between power consumption and processing time, deployment environment and data usage to achieve more accurate metrics to measure sustainability benefits, as well as problems in measuring social sustainability.